diff --git a/.gitattributes b/.gitattributes
index 4548baa58e32ea75a311e488d09f6ed8e86599d5..c4343a470fc7635b6a48a21921bfab8defcef5e0 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -12701,3 +12701,127 @@ SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/re
 SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/synthetic/pipeline_n16/real.csv filter=lfs diff=lfs merge=lfs -text
 SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/synthetic/pipeline_n16/test.csv filter=lfs diff=lfs merge=lfs -text
 SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/synthetic/pipeline_n16/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/gen_20260513_134510.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/models_tabbyflow/trained.pt filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/run_config.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/runtime_result.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow-n16-227845-20260513_134510.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow_train_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/info.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/real.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/train_20260513_131701.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260321_164225.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260424_212203.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_033728.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_080506.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_unnorm.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/loss.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model.pt filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model_ema.pt filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/y_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-1000-20260321_164225.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/train_20260321_164024.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/gen_20260501_194646.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/models_tabdiff/trained.pt filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/runtime_result.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff-n16-227845-20260501_194646.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff_train_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/info.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/real.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_val.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/train_20260501_191724.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/gen_20260512_082701.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/run_config.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/runtime_result.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/train_20260512_082701.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_test.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_train.npy filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/gen_20260427_002515.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/real.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/train_20260426_221025.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260328_164845.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260330_070842.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/train_20260328_053849.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-1000-20260328_164845.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv filter=lfs diff=lfs merge=lfs -text
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_config.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_config.py
new file mode 100644
index 0000000000000000000000000000000000000000..95c723f64388fe1259f0ba62f0ad8c6cf1051455
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_config.py
@@ -0,0 +1,62 @@
+import os
+from pathlib import Path
+
+from src.util import load_config
+from ef_vfm.modules.main_modules import UniModMLP
+
+
+CONFIG_PATH = Path(__file__).resolve().parent.parent / "ef_vfm" / "configs" / "ef_vfm_configs.toml"
+
+
+def test_load_config_returns_dict():
+    config = load_config(CONFIG_PATH)
+    assert isinstance(config, dict)
+
+
+def test_config_has_expected_sections():
+    config = load_config(CONFIG_PATH)
+    for key in ['data', 'unimodmlp_params', 'train', 'sample']:
+        assert key in config, f"Missing section '{key}'"
+
+
+def test_unimodmlp_params_complete():
+    config = load_config(CONFIG_PATH)
+    params = config['unimodmlp_params']
+    required = ['num_layers', 'd_token', 'n_head', 'factor', 'bias', 'dim_t', 'use_mlp', 'activation']
+    for key in required:
+        assert key in params, f"Missing param '{key}' in unimodmlp_params"
+
+
+def test_activation_value_is_valid():
+    config = load_config(CONFIG_PATH)
+    activation = config['unimodmlp_params']['activation']
+    assert activation in ('relu', 'gelu', 'silu'), f"Invalid activation '{activation}'"
+
+
+def test_train_main_has_new_params():
+    """Verify the recently added config params are present."""
+    config = load_config(CONFIG_PATH)
+    train = config['train']['main']
+    assert 'max_grad_norm' in train
+    assert 'warmup_epochs' in train
+    assert isinstance(train['max_grad_norm'], (int, float))
+    assert isinstance(train['warmup_epochs'], (int, float))
+
+
+def test_config_values_create_model():
+    config = load_config(CONFIG_PATH)
+    params = config['unimodmlp_params']
+    # Use dummy dimensions; the point is that config params are valid for the constructor
+    model = UniModMLP(
+        d_numerical=4,
+        categories=[3, 5, 2],
+        num_layers=params['num_layers'],
+        d_token=params['d_token'],
+        n_head=params['n_head'],
+        factor=params['factor'],
+        bias=params['bias'],
+        dim_t=params['dim_t'],
+        use_mlp=params['use_mlp'],
+        activation=params['activation'],
+    )
+    assert model is not None
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_flow_model.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_flow_model.py
new file mode 100644
index 0000000000000000000000000000000000000000..2bdc72bf2388cc70b56389463bdfd322b8badced
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_flow_model.py
@@ -0,0 +1,219 @@
+import torch
+import numpy as np
+from unittest.mock import patch
+
+from ef_vfm.models.flow_model import ExpVFM, Velocity
+from ef_vfm.modules.main_modules import UniModMLP
+
+
+# ---- mixed_loss tests ----
+
+def test_mixed_loss_returns_two_scalars(make_flow_model, make_dummy_inputs, dims):
+    d = dims
+    flow = make_flow_model(d["d_numerical"], d["categories"])
+    _, _, x_cat_int, _ = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num = torch.randn(d["batch_size"], d["d_numerical"])
+    x = torch.cat([x_num, x_cat_int.float()], dim=1)
+    d_loss, c_loss = flow.mixed_loss(x)
+    assert d_loss.dim() == 0 or d_loss.numel() == 1
+    assert c_loss.dim() == 0 or c_loss.numel() == 1
+
+
+def test_mixed_loss_finite(make_flow_model, make_dummy_inputs, dims):
+    d = dims
+    flow = make_flow_model(d["d_numerical"], d["categories"])
+    _, _, x_cat_int, _ = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num = torch.randn(d["batch_size"], d["d_numerical"])
+    x = torch.cat([x_num, x_cat_int.float()], dim=1)
+    d_loss, c_loss = flow.mixed_loss(x)
+    assert torch.isfinite(d_loss).all()
+    assert torch.isfinite(c_loss).all()
+
+
+def test_mixed_loss_gradients_flow(make_flow_model, make_dummy_inputs, dims):
+    d = dims
+    flow = make_flow_model(d["d_numerical"], d["categories"])
+    _, _, x_cat_int, _ = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num = torch.randn(d["batch_size"], d["d_numerical"])
+    x = torch.cat([x_num, x_cat_int.float()], dim=1)
+    d_loss, c_loss = flow.mixed_loss(x)
+    total = d_loss + c_loss
+    total.backward()
+    grads = [p.grad for p in flow.parameters() if p.grad is not None]
+    assert len(grads) > 0
+
+
+def test_mixed_loss_numerical_only(make_flow_model, make_dummy_inputs, dims_numerical_only):
+    d = dims_numerical_only
+    flow = make_flow_model(d["d_numerical"], d["categories"])
+    x = torch.randn(d["batch_size"], d["d_numerical"])
+    d_loss, c_loss = flow.mixed_loss(x)
+    assert d_loss.item() == 0.0  # no discrete features
+    assert c_loss.item() > 0.0
+
+
+# ---- sample tests (with mocked odeint) ----
+
+def _make_flow(d_numerical, categories):
+    cats_list = list(categories) if categories is not None else []
+    cats_np = np.array(cats_list)
+    model = UniModMLP(d_numerical, cats_list, 1, 16, n_head=1, factor=4, dim_t=64, activation='gelu')
+    return ExpVFM(cats_np, d_numerical, model, device=torch.device('cpu'))
+
+
+def test_sample_output_shape(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    d_in = d["d_numerical"] + sum(d["categories"])
+    n = 5
+    fake_trajectory = torch.randn(2, n, d_in)
+    with patch("ef_vfm.models.flow_model.odeint", return_value=fake_trajectory):
+        result = flow.sample(n)
+    d_out = d["d_numerical"] + len(d["categories"])
+    assert result.shape == (n, d_out)
+
+
+def test_sample_categorical_in_range(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    d_in = d["d_numerical"] + sum(d["categories"])
+    n = 16
+    fake_trajectory = torch.randn(2, n, d_in)
+    with patch("ef_vfm.models.flow_model.odeint", return_value=fake_trajectory):
+        result = flow.sample(n)
+    for i, k in enumerate(d["categories"]):
+        col = d["d_numerical"] + i
+        assert (result[:, col] >= 0).all()
+        assert (result[:, col] < k).all()
+
+
+def test_sample_returns_cpu(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    d_in = d["d_numerical"] + sum(d["categories"])
+    fake_trajectory = torch.randn(2, 4, d_in)
+    with patch("ef_vfm.models.flow_model.odeint", return_value=fake_trajectory):
+        result = flow.sample(4)
+    assert result.device == torch.device('cpu')
+
+
+def test_sample_single_sample(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    d_in = d["d_numerical"] + sum(d["categories"])
+    fake_trajectory = torch.randn(2, 1, d_in)
+    with patch("ef_vfm.models.flow_model.odeint", return_value=fake_trajectory):
+        result = flow.sample(1)
+    d_out = d["d_numerical"] + len(d["categories"])
+    assert result.shape == (1, d_out)
+
+
+# ---- to_one_hot tests ----
+
+def test_to_one_hot_shape(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    cats = d["categories"]
+    x_cat = torch.stack([torch.randint(0, k, (8,)) for k in cats], dim=1)
+    oh = flow.to_one_hot(x_cat)
+    assert oh.shape == (8, sum(cats))
+
+
+def test_to_one_hot_roundtrip(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    cats = d["categories"]
+    x_cat = torch.stack([torch.randint(0, k, (8,)) for k in cats], dim=1)
+    oh = flow.to_one_hot(x_cat)
+    # Recover indices via argmax per category slice
+    idx = 0
+    for i, k in enumerate(cats):
+        recovered = oh[:, idx:idx + k].argmax(dim=1)
+        assert torch.equal(recovered, x_cat[:, i])
+        idx += k
+
+
+def test_to_one_hot_binary_values(dims):
+    d = dims
+    flow = _make_flow(d["d_numerical"], d["categories"])
+    cats = d["categories"]
+    x_cat = torch.stack([torch.randint(0, k, (8,)) for k in cats], dim=1)
+    oh = flow.to_one_hot(x_cat)
+    assert set(oh.unique().tolist()).issubset({0, 1})
+
+
+# ---- Regression tests ----
+
+def test_regression_d_in_no_extra_len():
+    """d_in must be num_numerical + sum(num_classes), NOT + len(num_classes)."""
+    d_numerical = 4
+    categories = np.array([3, 5, 2])
+    flow = _make_flow(d_numerical, categories)
+    expected_d_in = d_numerical + sum(categories)  # 14, not 17
+    assert flow.num_numerical_features + sum(flow.num_classes) == expected_d_in
+
+
+def test_regression_sampling_indices_correct():
+    """Categorical argmax must go to columns [d_num, d_num+1, ...], not [0, 1, ...]."""
+    d_numerical = 4
+    categories = np.array([3, 5, 2])
+    n = 10
+    d_in = d_numerical + sum(categories)
+    d_out = d_numerical + len(categories)
+
+    # Simulate the post-processing from sample()
+    out = torch.randn(n, d_in)
+    sample = torch.zeros(n, d_out)
+    sample[:, :d_numerical] = out[:, :d_numerical]
+
+    idx = d_numerical  # correct starting index
+    for i, val in enumerate(categories):
+        col = d_numerical + i  # correct column
+        sample[:, col] = torch.argmax(out[:, idx:idx + val], dim=1)
+        idx += val
+
+    # Numerical columns must be untouched
+    assert torch.allclose(sample[:, :d_numerical], out[:, :d_numerical])
+    # Categorical columns at correct positions
+    for i, val in enumerate(categories):
+        col = d_numerical + i
+        assert (sample[:, col] >= 0).all()
+        assert (sample[:, col] < val).all()
+
+
+def test_regression_d_out_correct():
+    """d_out must be d_num + len(categories)."""
+    d_numerical = 4
+    categories = np.array([3, 5, 2])
+    flow = _make_flow(d_numerical, categories)
+    expected_d_out = d_numerical + len(categories)  # 7
+    assert expected_d_out == 7
+
+
+# ---- Velocity tests ----
+
+def test_velocity_output_shape(dims):
+    d = dims
+    cats_list = list(d["categories"])
+    model = UniModMLP(d["d_numerical"], cats_list, 1, d["d_token"],
+                      n_head=1, factor=4, dim_t=64, activation='gelu')
+    vel = Velocity(model)
+    d_in = d["d_numerical"] + sum(d["categories"])
+    x = torch.randn(d["batch_size"], d_in)
+    t = torch.tensor(0.5)
+    out = vel(t, x)
+    assert out.shape == (d["batch_size"], d_in)
+
+
+def test_velocity_scalar_t_broadcast(dims):
+    d = dims
+    cats_list = list(d["categories"])
+    model = UniModMLP(d["d_numerical"], cats_list, 1, d["d_token"],
+                      n_head=1, factor=4, dim_t=64, activation='gelu')
+    vel = Velocity(model)
+    d_in = d["d_numerical"] + sum(d["categories"])
+    x = torch.randn(d["batch_size"], d_in)
+    # Scalar t should work (gets broadcast internally)
+    t = torch.tensor(0.3)
+    out = vel(t, x)
+    assert out.shape == x.shape
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_mlp.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_mlp.py
new file mode 100644
index 0000000000000000000000000000000000000000..0cf9ad4d6832d792cb65a1bb01bdc784385f9fcd
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_mlp.py
@@ -0,0 +1,85 @@
+import torch
+import torch.nn as nn
+from ef_vfm.modules.main_modules import MLP, PositionalEmbedding
+
+
+# ---- PositionalEmbedding tests ----
+
+def test_positional_embedding_shape():
+    pe = PositionalEmbedding(num_channels=64)
+    x = torch.rand(8)
+    out = pe(x)
+    assert out.shape == (8, 64)
+
+
+def test_positional_embedding_bounded():
+    pe = PositionalEmbedding(num_channels=64)
+    x = torch.rand(8)
+    out = pe(x)
+    assert out.min() >= -1.0
+    assert out.max() <= 1.0
+
+
+def test_positional_embedding_deterministic():
+    pe = PositionalEmbedding(num_channels=64)
+    x = torch.tensor([0.1, 0.5, 0.9])
+    out1 = pe(x)
+    out2 = pe(x)
+    assert torch.equal(out1, out2)
+
+
+def test_positional_embedding_different_timesteps():
+    pe = PositionalEmbedding(num_channels=64)
+    t1 = torch.tensor([0.1])
+    t2 = torch.tensor([0.9])
+    assert not torch.allclose(pe(t1), pe(t2))
+
+
+# ---- MLP tests ----
+
+def test_mlp_output_shape(make_mlp):
+    mlp = make_mlp(d_in=32, dim_t=64)
+    x = torch.randn(8, 32)
+    t = torch.rand(8)
+    out = mlp(x, t)
+    assert out.shape == (8, 32)
+
+
+def test_mlp_use_mlp_true(make_mlp):
+    mlp = make_mlp(d_in=32, dim_t=64, use_mlp=True)
+    assert isinstance(mlp.mlp, nn.Sequential)
+
+
+def test_mlp_use_mlp_false(make_mlp):
+    mlp = make_mlp(d_in=32, dim_t=64, use_mlp=False)
+    assert isinstance(mlp.mlp, nn.Linear)
+
+
+def test_mlp_time_conditioning(make_mlp):
+    mlp = make_mlp(d_in=32, dim_t=64)
+    mlp.eval()
+    x = torch.randn(4, 32)
+    t1 = torch.zeros(4)
+    t2 = torch.ones(4)
+    out1 = mlp(x, t1)
+    out2 = mlp(x, t2)
+    assert not torch.allclose(out1, out2)
+
+
+def test_mlp_gradient_flows(make_mlp):
+    mlp = make_mlp(d_in=32, dim_t=64)
+    x = torch.randn(4, 32)
+    t = torch.rand(4)
+    out = mlp(x, t)
+    out.sum().backward()
+    assert mlp.proj.weight.grad is not None and mlp.proj.weight.grad.abs().sum() > 0
+    assert mlp.map_noise.num_channels == 64  # sanity check on PE config
+
+
+def test_mlp_different_dim_t(make_mlp):
+    for dim_t in [32, 128, 256]:
+        mlp = make_mlp(d_in=16, dim_t=dim_t)
+        x = torch.randn(4, 16)
+        t = torch.rand(4)
+        out = mlp(x, t)
+        assert out.shape == (4, 16)
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_reconstructor.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_reconstructor.py
new file mode 100644
index 0000000000000000000000000000000000000000..cdc39880d19f644fb8ac6b457af6a8cb3d83cbfa
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_reconstructor.py
@@ -0,0 +1,51 @@
+import torch
+import numpy as np
+from ef_vfm.modules.transformer import Reconstructor
+
+
+def test_output_shapes_mixed(make_reconstructor, dims):
+    d = dims
+    r = make_reconstructor(d["d_numerical"], d["categories"], d["d_token"])
+    seq_len = d["d_numerical"] + len(d["categories"])
+    h = torch.randn(d["batch_size"], seq_len, d["d_token"])
+    x_num, x_cat = r(h)
+    assert x_num.shape == (d["batch_size"], d["d_numerical"])
+    assert len(x_cat) == len(d["categories"])
+    for i, k in enumerate(d["categories"]):
+        assert x_cat[i].shape == (d["batch_size"], k)
+
+
+def test_categorical_count(make_reconstructor, dims):
+    d = dims
+    r = make_reconstructor(d["d_numerical"], d["categories"], d["d_token"])
+    seq_len = d["d_numerical"] + len(d["categories"])
+    h = torch.randn(d["batch_size"], seq_len, d["d_token"])
+    _, x_cat = r(h)
+    assert len(x_cat) == len(d["categories"])
+
+
+def test_empty_categories(make_reconstructor):
+    r = make_reconstructor(4, np.array([]), 16)
+    h = torch.randn(8, 4, 16)
+    x_num, x_cat = r(h)
+    assert x_num.shape == (8, 4)
+    assert len(x_cat) == 0
+
+
+def test_weight_shape(make_reconstructor, dims):
+    d = dims
+    r = make_reconstructor(d["d_numerical"], d["categories"], d["d_token"])
+    assert r.weight.shape == (d["d_numerical"], d["d_token"])
+
+
+def test_gradient_flows(make_reconstructor, dims):
+    d = dims
+    r = make_reconstructor(d["d_numerical"], d["categories"], d["d_token"])
+    seq_len = d["d_numerical"] + len(d["categories"])
+    h = torch.randn(d["batch_size"], seq_len, d["d_token"])
+    x_num, x_cat = r(h)
+    loss = x_num.sum() + sum(c.sum() for c in x_cat)
+    loss.backward()
+    assert r.weight.grad is not None and r.weight.grad.abs().sum() > 0
+    for recon in r.cat_recons:
+        assert recon.weight.grad is not None
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_tokenizer.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_tokenizer.py
new file mode 100644
index 0000000000000000000000000000000000000000..ea8c55737d473605c2bb1c87f0394fad64baeb18
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_tokenizer.py
@@ -0,0 +1,85 @@
+import torch
+import numpy as np
+
+
+def test_forward_shape_mixed(make_tokenizer, make_dummy_inputs, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"])
+    x_num, x_cat_oh, _, _ = make_dummy_inputs(dims["d_numerical"], dims["categories"], dims["batch_size"])
+    out = tok(x_num, x_cat_oh)
+    expected_seq = 1 + dims["d_numerical"] + len(dims["categories"])
+    assert out.shape == (dims["batch_size"], expected_seq, dims["d_token"])
+
+
+def test_forward_shape_numerical_only(make_tokenizer, make_dummy_inputs, dims_numerical_only):
+    d = dims_numerical_only
+    tok = make_tokenizer(d["d_numerical"], d["categories"], d["d_token"])
+    x_num, _, _, _ = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    out = tok(x_num, None)
+    expected_seq = 1 + d["d_numerical"]
+    assert out.shape == (d["batch_size"], expected_seq, d["d_token"])
+
+
+def test_forward_shape_single_feature(make_tokenizer, make_dummy_inputs, dims_single):
+    d = dims_single
+    tok = make_tokenizer(d["d_numerical"], d["categories"], d["d_token"])
+    x_num, x_cat_oh, _, _ = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    out = tok(x_num, x_cat_oh)
+    expected_seq = 1 + d["d_numerical"] + len(d["categories"])
+    assert out.shape == (d["batch_size"], expected_seq, d["d_token"])
+
+
+def test_n_tokens_property(make_tokenizer, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"])
+    expected = dims["d_numerical"] + 1 + len(dims["categories"])
+    assert tok.n_tokens == expected
+
+
+def test_n_tokens_numerical_only(make_tokenizer, dims_numerical_only):
+    d = dims_numerical_only
+    tok = make_tokenizer(d["d_numerical"], d["categories"], d["d_token"])
+    assert tok.n_tokens == d["d_numerical"] + 1
+
+
+def test_cls_token_position(make_tokenizer, make_dummy_inputs, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"], bias=False)
+    x_num, x_cat_oh, _, _ = make_dummy_inputs(dims["d_numerical"], dims["categories"], dims["batch_size"])
+    out = tok(x_num, x_cat_oh)
+    # CLS token: ones * weight[0], so all batch rows should have the same CLS token
+    cls_tokens = out[:, 0, :]
+    assert torch.allclose(cls_tokens[0], cls_tokens[1])
+    assert torch.allclose(cls_tokens[0], tok.weight[0])
+
+
+def test_bias_vs_no_bias(make_tokenizer, make_dummy_inputs, dims):
+    d = dims
+    tok_bias = make_tokenizer(d["d_numerical"], d["categories"], d["d_token"], bias=True)
+    tok_no_bias = make_tokenizer(d["d_numerical"], d["categories"], d["d_token"], bias=False)
+    assert tok_bias.bias is not None
+    assert tok_no_bias.bias is None
+
+
+def test_category_offsets_values(make_tokenizer):
+    cats = np.array([3, 5, 2])
+    tok = make_tokenizer(4, cats, 16)
+    assert torch.equal(tok.category_offsets, torch.tensor([0, 3, 8]))
+    assert torch.equal(tok.category_ends, torch.tensor([3, 8, 10]))
+
+
+def test_cat_weight_shape(make_tokenizer, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"])
+    assert tok.cat_weight.shape == (sum(dims["categories"]), dims["d_token"])
+
+
+def test_weight_shape(make_tokenizer, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"])
+    assert tok.weight.shape == (dims["d_numerical"] + 1, dims["d_token"])
+
+
+def test_gradient_flows(make_tokenizer, make_dummy_inputs, dims):
+    tok = make_tokenizer(dims["d_numerical"], dims["categories"], dims["d_token"])
+    x_num, x_cat_oh, _, _ = make_dummy_inputs(dims["d_numerical"], dims["categories"], dims["batch_size"])
+    out = tok(x_num, x_cat_oh)
+    out.sum().backward()
+    assert tok.weight.grad is not None and tok.weight.grad.abs().sum() > 0
+    assert tok.cat_weight.grad is not None and tok.cat_weight.grad.abs().sum() > 0
+    assert tok.bias.grad is not None and tok.bias.grad.abs().sum() > 0
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_trainer.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_trainer.py
new file mode 100644
index 0000000000000000000000000000000000000000..592c538d2aa1f8f34098a0d7c6c4fc0b1f0c5ddf
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_trainer.py
@@ -0,0 +1,98 @@
+import torch
+import numpy as np
+
+
+# ---- Gradient clipping tests ----
+
+def test_grad_clipping_applied(make_trainer, tmp_path):
+    trainer = make_trainer(max_grad_norm=0.5, tmp_path=tmp_path)
+    batch = next(iter(trainer.train_iter))
+    trainer._run_step(batch, closs_weight=1.0, dloss_weight=1.0)
+    # After clipping, total gradient norm should be <= max_grad_norm (with tolerance)
+    total_norm = torch.nn.utils.clip_grad_norm_(trainer.flow.parameters(), float('inf'))
+    # Gradients were already clipped in _run_step, then optimizer.step() zeroed them.
+    # So we re-run to check: do a fresh forward-backward without step
+    trainer.optimizer.zero_grad()
+    dloss, closs = trainer.flow.mixed_loss(batch.to(trainer.device))
+    (dloss + closs).backward()
+    torch.nn.utils.clip_grad_norm_(trainer.flow.parameters(), 0.5)
+    total_norm = 0.0
+    for p in trainer.flow.parameters():
+        if p.grad is not None:
+            total_norm += p.grad.data.norm(2).item() ** 2
+    total_norm = total_norm ** 0.5
+    assert total_norm <= 0.5 + 1e-6
+
+
+def test_grad_clipping_disabled(make_trainer, tmp_path):
+    trainer = make_trainer(max_grad_norm=0, tmp_path=tmp_path)
+    assert trainer.max_grad_norm == 0
+
+
+def test_run_step_returns_losses(make_trainer, tmp_path):
+    trainer = make_trainer(tmp_path=tmp_path)
+    batch = next(iter(trainer.train_iter))
+    dloss, closs = trainer._run_step(batch, closs_weight=1.0, dloss_weight=1.0)
+    assert isinstance(dloss, torch.Tensor)
+    assert isinstance(closs, torch.Tensor)
+    assert torch.isfinite(dloss)
+    assert torch.isfinite(closs)
+
+
+# ---- LR warmup tests ----
+
+def test_warmup_lr_linear_ramp(make_trainer, tmp_path):
+    init_lr = 0.01
+    warmup = 5
+    trainer = make_trainer(lr=init_lr, warmup_epochs=warmup, tmp_path=tmp_path)
+    # Simulate warmup epochs
+    for epoch in range(warmup):
+        expected_lr = init_lr * (epoch + 1) / warmup
+        if trainer.warmup_epochs > 0 and (epoch + 1) <= trainer.warmup_epochs:
+            warmup_lr = trainer.init_lr * (epoch + 1) / trainer.warmup_epochs
+            for pg in trainer.optimizer.param_groups:
+                pg["lr"] = warmup_lr
+        actual_lr = trainer.optimizer.param_groups[0]["lr"]
+        assert abs(actual_lr - expected_lr) < 1e-8, f"Epoch {epoch}: expected {expected_lr}, got {actual_lr}"
+
+
+def test_warmup_overrides_scheduler(make_trainer, tmp_path):
+    trainer = make_trainer(warmup_epochs=10, lr_scheduler='reduce_lr_on_plateau', tmp_path=tmp_path)
+    initial_lr = trainer.optimizer.param_groups[0]["lr"]
+    # During warmup, scheduler.step should NOT be called (we just set LR directly)
+    # Simulate epoch 1 warmup
+    warmup_lr = trainer.init_lr * 1 / trainer.warmup_epochs
+    for pg in trainer.optimizer.param_groups:
+        pg["lr"] = warmup_lr
+    assert trainer.optimizer.param_groups[0]["lr"] == warmup_lr
+    assert warmup_lr < initial_lr  # warmup starts lower
+
+
+def test_no_warmup_when_zero(make_trainer, tmp_path):
+    trainer = make_trainer(warmup_epochs=0, tmp_path=tmp_path)
+    assert trainer.warmup_epochs == 0
+    # LR should be the init_lr from the start
+    assert trainer.optimizer.param_groups[0]["lr"] == trainer.init_lr
+
+
+# ---- LR scheduler tests ----
+
+def test_anneal_lr(make_trainer, tmp_path):
+    trainer = make_trainer(lr=0.01, steps=100, lr_scheduler='anneal', tmp_path=tmp_path)
+    trainer._anneal_lr(50)
+    expected = 0.01 * (1 - 50 / 100)
+    assert abs(trainer.optimizer.param_groups[0]["lr"] - expected) < 1e-8
+
+
+# ---- EMA tests ----
+
+def test_ema_model_created(make_trainer, tmp_path):
+    trainer = make_trainer(tmp_path=tmp_path)
+    # EMA model should exist and have same structure as flow._vf_fn
+    assert trainer.ema_model is not None
+    ema_params = list(trainer.ema_model.parameters())
+    model_params = list(trainer.flow._vf_fn.parameters())
+    assert len(ema_params) == len(model_params)
+    # EMA params should be detached (requires_grad=False)
+    for p in ema_params:
+        assert not p.requires_grad
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_transformer.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_transformer.py
new file mode 100644
index 0000000000000000000000000000000000000000..ff56e884615e818841fbb912f8ef4e9961729197
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_transformer.py
@@ -0,0 +1,73 @@
+import pytest
+import torch
+from ef_vfm.modules.transformer import Transformer
+
+
+def test_output_shape_preserved(make_transformer):
+    t = make_transformer(d_token=16, n_layers=2)
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_activation_gelu(make_transformer):
+    t = make_transformer(d_token=16, activation='gelu')
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_activation_silu(make_transformer):
+    t = make_transformer(d_token=16, activation='silu')
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_activation_relu(make_transformer):
+    t = make_transformer(d_token=16, activation='relu')
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_invalid_activation_raises():
+    with pytest.raises(ValueError, match="Unknown activation"):
+        Transformer(2, 16, 1, 16, 4, activation='bad')
+
+
+def test_prenorm_first_layer_no_norm0():
+    t = Transformer(2, 16, 1, 16, 4, prenormalization=True)
+    assert 'norm0' not in t.layers[0]
+    # Second layer should have norm0
+    assert 'norm0' in t.layers[1]
+
+
+def test_no_prenorm_all_layers_have_norm0():
+    t = Transformer(2, 16, 1, 16, 4, prenormalization=False)
+    for layer in t.layers:
+        assert 'norm0' in layer
+
+
+def test_single_layer():
+    t = Transformer(1, 16, 1, 16, 4)
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_multi_layer():
+    t = Transformer(4, 16, 1, 16, 4)
+    x = torch.randn(4, 5, 16)
+    out = t(x)
+    assert out.shape == x.shape
+
+
+def test_gradient_flows(make_transformer):
+    t = make_transformer(d_token=16, n_layers=2)
+    x = torch.randn(4, 5, 16, requires_grad=True)
+    out = t(x)
+    out.sum().backward()
+    assert x.grad is not None and x.grad.abs().sum() > 0
+    # Check gradients through at least the first layer's linear0
+    assert t.layers[0]['linear0'].weight.grad is not None
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_unimodmlp.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_unimodmlp.py
new file mode 100644
index 0000000000000000000000000000000000000000..d935e7c48dba78fd7e4821287628aa861aa6d1b4
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_unimodmlp.py
@@ -0,0 +1,72 @@
+import torch
+import numpy as np
+
+
+def test_forward_shapes_mixed(make_unimodmlp, make_dummy_inputs, dims):
+    d = dims
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    x_num, x_cat_oh, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num_pred, x_cat_pred = model(x_num, x_cat_oh, t)
+    assert x_num_pred.shape == (d["batch_size"], d["d_numerical"])
+    assert x_cat_pred.shape == (d["batch_size"], sum(d["categories"]))
+
+
+def test_forward_shapes_numerical_only(make_unimodmlp, make_dummy_inputs, dims_numerical_only):
+    d = dims_numerical_only
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    x_num, _, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_cat = torch.zeros(d["batch_size"], 0)
+    x_num_pred, x_cat_pred = model(x_num, x_cat, t)
+    assert x_num_pred.shape == (d["batch_size"], d["d_numerical"])
+    # When no categories, cat_pred should be zeros with shape matching x_cat
+    assert x_cat_pred.shape[0] == d["batch_size"]
+    assert torch.all(x_cat_pred == 0)
+
+
+def test_forward_shapes_single_feature(make_unimodmlp, make_dummy_inputs, dims_single):
+    d = dims_single
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    x_num, x_cat_oh, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num_pred, x_cat_pred = model(x_num, x_cat_oh, t)
+    assert x_num_pred.shape == (d["batch_size"], d["d_numerical"])
+    assert x_cat_pred.shape == (d["batch_size"], sum(d["categories"]))
+
+
+def test_d_in_computation(make_unimodmlp, dims):
+    d = dims
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    expected = d["d_token"] * (d["d_numerical"] + len(d["categories"]))
+    assert model.mlp.proj.in_features == expected
+
+
+def test_output_dtypes(make_unimodmlp, make_dummy_inputs, dims):
+    d = dims
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    x_num, x_cat_oh, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num_pred, x_cat_pred = model(x_num, x_cat_oh, t)
+    assert x_num_pred.dtype == torch.float32
+    assert x_cat_pred.dtype == torch.float32
+
+
+def test_gradient_flows_end_to_end(make_unimodmlp, make_dummy_inputs, dims):
+    d = dims
+    model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"])
+    x_num, x_cat_oh, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    x_num_pred, x_cat_pred = model(x_num, x_cat_oh, t)
+    loss = x_num_pred.sum() + x_cat_pred.sum()
+    loss.backward()
+    params_with_grad = sum(1 for p in model.parameters() if p.grad is not None and p.grad.abs().sum() > 0)
+    total_params = sum(1 for _ in model.parameters())
+    # Transformer.head is defined but unused in forward(), so not all params get gradients
+    assert params_with_grad > total_params * 0.8, f"Only {params_with_grad}/{total_params} params got gradients"
+
+
+def test_different_activations(make_unimodmlp, make_dummy_inputs, dims):
+    d = dims
+    x_num, x_cat_oh, _, t = make_dummy_inputs(d["d_numerical"], d["categories"], d["batch_size"])
+    for act in ['relu', 'gelu', 'silu']:
+        model = make_unimodmlp(d["d_numerical"], d["categories"], d_token=d["d_token"], activation=act)
+        x_num_pred, x_cat_pred = model(x_num, x_cat_oh, t)
+        assert x_num_pred.shape == (d["batch_size"], d["d_numerical"])
+        assert torch.isfinite(x_num_pred).all()
+        assert torch.isfinite(x_cat_pred).all()
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_utils.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..fbce0db0ac74052a4038dee89fd753b9d97717aa
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/tests/test_utils.py
@@ -0,0 +1,49 @@
+import torch
+import numpy as np
+
+from utils_train import update_ema, concat_y_to_X
+
+
+# ---- update_ema tests ----
+
+def test_update_ema_basic():
+    target = [torch.tensor([1.0, 2.0])]
+    source = [torch.tensor([3.0, 4.0])]
+    target[0].requires_grad_(False)
+    rate = 0.9
+    update_ema(target, source, rate=rate)
+    expected = 0.9 * torch.tensor([1.0, 2.0]) + 0.1 * torch.tensor([3.0, 4.0])
+    assert torch.allclose(target[0], expected)
+
+
+def test_update_ema_rate_zero():
+    target = [torch.tensor([1.0, 2.0])]
+    source = [torch.tensor([3.0, 4.0])]
+    target[0].requires_grad_(False)
+    update_ema(target, source, rate=0.0)
+    assert torch.allclose(target[0], torch.tensor([3.0, 4.0]))
+
+
+def test_update_ema_rate_one():
+    target = [torch.tensor([1.0, 2.0])]
+    source = [torch.tensor([3.0, 4.0])]
+    target[0].requires_grad_(False)
+    update_ema(target, source, rate=1.0)
+    assert torch.allclose(target[0], torch.tensor([1.0, 2.0]))
+
+
+# ---- concat_y_to_X tests ----
+
+def test_concat_y_to_X_with_X():
+    X = np.array([[1, 2], [3, 4]])
+    y = np.array([10, 20])
+    result = concat_y_to_X(X, y)
+    expected = np.array([[10, 1, 2], [20, 3, 4]])
+    np.testing.assert_array_equal(result, expected)
+
+
+def test_concat_y_to_X_without_X():
+    y = np.array([10, 20, 30])
+    result = concat_y_to_X(None, y)
+    expected = np.array([[10], [20], [30]])
+    np.testing.assert_array_equal(result, expected)
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/utils_train.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/utils_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..f849c412fa3cc62098f14b12684d16ecc30e94ea
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/utils_train.py
@@ -0,0 +1,205 @@
+import numpy as np
+import os
+from pathlib import Path
+
+import src
+from torch.utils.data import Dataset
+
+import torch
+
+
+class TabularDataset(Dataset):
+    def __init__(self, X_num, X_cat):
+        self.X_num = X_num
+        self.X_cat = X_cat
+
+    def __getitem__(self, index):
+        this_num = self.X_num[index]
+        this_cat = self.X_cat[index]
+
+        sample = (this_num, this_cat)
+
+        return sample
+
+    def __len__(self):
+        return self.X_num.shape[0]
+
+
+class EFVFMDataset(Dataset):
+    def __init__(self, dataname, data_dir, info, isTrain=True, dequant_dist='none', int_dequant_factor=0.0):
+        self.dataname = dataname
+        self.data_dir = data_dir
+        self.info = info
+        self.isTrain = isTrain
+
+        X_num, X_cat, categories, d_numerical, num_inverse, int_inverse, cat_inverse = preprocess(
+            data_dir, dequant_dist, int_dequant_factor, task_type=info['task_type'], inverse=True
+        )
+        categories = np.array(categories)
+
+        X_train_num, X_test_num = X_num
+        X_train_cat, X_test_cat = X_cat
+
+        X_train_num = torch.tensor(X_train_num).float()
+        X_test_num = torch.tensor(X_test_num).float()
+        X_train_cat = torch.tensor(X_train_cat)
+        X_test_cat = torch.tensor(X_test_cat)
+
+        self.X = (
+            torch.cat((X_train_num, X_train_cat), dim=1)
+            if isTrain
+            else torch.cat((X_test_num, X_test_cat), dim=1)
+        )
+        self.num_inverse = num_inverse
+        self.int_inverse = int_inverse
+        self.cat_inverse = cat_inverse
+        self.d_numerical = d_numerical
+        self.categories = categories
+
+    def __getitem__(self, index):
+        return self.X[index]
+
+    def __len__(self):
+        return self.X.shape[0]
+
+
+def _empty_num_like(y_split):
+    return np.zeros((len(y_split), 0), dtype=np.float32)
+
+
+def _empty_cat_like(y_split):
+    return np.zeros((len(y_split), 0), dtype=np.int64)
+
+
+def preprocess(dataset_path, dequant_dist='none', int_dequant_factor=0.0, task_type='binclass', inverse=False, cat_encoding=None, concat=True):
+
+    T_dict = {}
+
+    T_dict['normalization'] = "quantile"
+    T_dict['num_nan_policy'] = 'mean'
+    T_dict['cat_nan_policy'] = None
+    T_dict['cat_min_frequency'] = None
+    T_dict['cat_encoding'] = cat_encoding
+    T_dict['y_policy'] = "default"
+    T_dict['dequant_dist'] = dequant_dist
+    T_dict['int_dequant_factor'] = int_dequant_factor
+
+    T = src.Transformations(**T_dict)
+
+    dataset = make_dataset(
+        data_path=dataset_path,
+        T=T,
+        task_type=task_type,
+        change_val=False,
+        concat=concat,
+    )
+
+    if cat_encoding is None:
+        X_num = dataset.X_num
+        X_cat = dataset.X_cat
+        y = dataset.y
+
+        if X_num is None:
+            X_train_num = _empty_num_like(y['train'])
+            X_test_num = _empty_num_like(y['test'])
+        else:
+            X_train_num, X_test_num = X_num['train'], X_num['test']
+
+        if X_cat is None:
+            # Some datasets have no categorical features after preprocessing.
+            # For classification tasks, ef-vfm still expects the target to be
+            # concatenated into the categorical block.
+            if task_type in ('binclass', 'multiclass') and concat and y is not None:
+                X_train_cat = y['train'].reshape(-1, 1)
+                X_test_cat = y['test'].reshape(-1, 1)
+            else:
+                X_train_cat = _empty_cat_like(y['train'])
+                X_test_cat = _empty_cat_like(y['test'])
+        else:
+            X_train_cat, X_test_cat = X_cat['train'], X_cat['test']
+
+        categories = src.get_categories(X_train_cat) if X_train_cat.shape[1] > 0 else []
+        d_numerical = X_train_num.shape[1]
+
+        X_num = (X_train_num, X_test_num)
+        X_cat = (X_train_cat, X_test_cat)
+
+        if inverse:
+            num_inverse = dataset.num_transform.inverse_transform if dataset.num_transform is not None else lambda x: x
+            int_inverse = dataset.int_transform.inverse_transform if dataset.int_transform is not None else lambda x: x
+            cat_inverse = dataset.cat_transform.inverse_transform if dataset.cat_transform is not None else lambda x: x
+
+            return X_num, X_cat, categories, d_numerical, num_inverse, int_inverse, cat_inverse
+        else:
+            return X_num, X_cat, categories, d_numerical
+    else:
+        return dataset
+
+
+def update_ema(target_params, source_params, rate=0.999):
+    for target, source in zip(target_params, source_params):
+        target.detach().mul_(rate).add_(source.detach(), alpha=1 - rate)
+
+
+def concat_y_to_X(X, y):
+    if X is None:
+        return y.reshape(-1, 1)
+    return np.concatenate([y.reshape(-1, 1), X], axis=1)
+
+
+def make_dataset(
+    data_path: str,
+    T: src.Transformations,
+    task_type,
+    change_val: bool,
+    concat=True,
+):
+
+    if task_type == 'binclass' or task_type == 'multiclass':
+        X_cat = {} if os.path.exists(os.path.join(data_path, 'X_cat_train.npy')) else None
+        X_num = {} if os.path.exists(os.path.join(data_path, 'X_num_train.npy')) else None
+        y = {} if os.path.exists(os.path.join(data_path, 'y_train.npy')) else None
+
+        for split in ['train', 'test']:
+            X_num_t, X_cat_t, y_t = src.read_pure_data(data_path, split)
+            if X_num is not None:
+                X_num[split] = X_num_t
+            if X_cat is not None:
+                if concat:
+                    X_cat_t = concat_y_to_X(X_cat_t, y_t)
+                X_cat[split] = X_cat_t
+            if y is not None:
+                y[split] = y_t
+    else:
+        X_cat = {} if os.path.exists(os.path.join(data_path, 'X_cat_train.npy')) else None
+        X_num = {} if os.path.exists(os.path.join(data_path, 'X_num_train.npy')) else None
+        y = {} if os.path.exists(os.path.join(data_path, 'y_train.npy')) else None
+
+        for split in ['train', 'test']:
+            X_num_t, X_cat_t, y_t = src.read_pure_data(data_path, split)
+            if X_num is not None:
+                if concat:
+                    X_num_t = concat_y_to_X(X_num_t, y_t)
+                X_num[split] = X_num_t
+            if X_cat is not None:
+                X_cat[split] = X_cat_t
+            if y is not None:
+                y[split] = y_t
+
+    info = src.load_json(os.path.join(data_path, 'info.json'))
+    int_col_idx_wrt_num = info['int_col_idx_wrt_num']
+
+    D = src.Dataset(
+        X_num,
+        X_cat,
+        y,
+        int_col_idx_wrt_num,
+        y_info={},
+        task_type=src.TaskType(info['task_type']),
+        n_classes=info.get('n_classes')
+    )
+
+    if change_val:
+        D = src.change_val(D)
+    D = src.transform_dataset(D, T, cache_dir=Path(data_path))
+    return D
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_gen.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_gen.py
new file mode 100644
index 0000000000000000000000000000000000000000..a1e5463d005350f968efc006f04489bfc15c3f4c
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_gen.py
@@ -0,0 +1,51 @@
+
+import os, shutil, subprocess, sys
+root = r"/workspace/ef-vfm"
+rt = r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime"
+name = r"pipeline_n16"
+src = r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16"
+
+if not os.path.exists(rt):
+    def _ignore(_, names):
+        skip = {"__pycache__", "data", "synthetic", "result", "results", "ckpt"}
+        return [n for n in names if n in skip or n.endswith(".pyc")]
+    shutil.copytree(root, rt, ignore=_ignore)
+
+dst_data = os.path.join(rt, "data", name)
+shutil.rmtree(dst_data, ignore_errors=True)
+os.makedirs(os.path.dirname(dst_data), exist_ok=True)
+shutil.copytree(src, dst_data)
+dst_syn = os.path.join(rt, "synthetic", name)
+os.makedirs(dst_syn, exist_ok=True)
+for fn in ("real.csv", "test.csv", "val.csv"):
+    shutil.copy(os.path.join(src, fn), os.path.join(dst_syn, fn))
+os.chdir(rt)
+os.environ["PYTHONPATH"] = rt + os.pathsep + os.environ.get("PYTHONPATH", "")
+os.environ.setdefault("EFVFM_SAMPLE_BATCH_SIZE", "64")
+os.environ.setdefault("EFVFM_ODE_FALLBACK", "1")
+os.environ.setdefault("EFVFM_RK4_STEPS", "32")
+subprocess.check_call([
+    sys.executable, os.path.join(rt, "main.py"),
+    "--dataname", name, "--mode", "test", "--gpu", "0",
+    "--no_wandb", "--exp_name", r"adapter_efvfm",
+    "--ckpt_path", r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime/ckpt/pipeline_n16/adapter_efvfm/model_90.pt",
+    "--num_samples_to_generate", str(int(227845)),
+])
+search_roots = [
+    os.path.join(rt, "result", name, r"adapter_efvfm"),
+    os.path.join(rt, "ef_vfm", "result", name, r"adapter_efvfm"),
+]
+best = None
+best_t = -1.0
+for base in search_roots:
+    if not os.path.isdir(base):
+        continue
+    for r, _, files in os.walk(base):
+        if "samples.csv" in files:
+            p = os.path.join(r, "samples.csv")
+            t = os.path.getmtime(p)
+            if t > best_t:
+                best_t, best = t, p
+if not best:
+    raise SystemExit("tabbyflow: no samples.csv in " + " | ".join(search_roots))
+shutil.copy(best, r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow-n16-227845-20260513_134510.csv")
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_train.py b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..47903ce43417c33bfe268c84fade11aeb5dc631f
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/_tabbyflow_train.py
@@ -0,0 +1,40 @@
+
+import os, shutil, subprocess, sys
+root = r"/workspace/ef-vfm"
+rt = r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/_efvfm_runtime"
+name = r"pipeline_n16"
+src = r"/work/output-Benchmark-trainonly-v1/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16"
+
+shutil.rmtree(rt, ignore_errors=True)
+
+def _ignore(_, names):
+    skip = {"__pycache__", "data", "synthetic", "result", "results", "ckpt"}
+    return [n for n in names if n in skip or n.endswith(".pyc")]
+
+shutil.copytree(root, rt, ignore=_ignore)
+pkg_cfg = os.path.join(rt, "ef_vfm", "configs")
+root_cfg = os.path.join(rt, "configs")
+if not os.path.isdir(root_cfg) and os.path.isdir(pkg_cfg):
+    shutil.copytree(pkg_cfg, root_cfg)
+dst_data = os.path.join(rt, "data", name)
+dst_syn = os.path.join(rt, "synthetic", name)
+shutil.rmtree(dst_data, ignore_errors=True)
+os.makedirs(os.path.dirname(dst_data), exist_ok=True)
+shutil.copytree(src, dst_data)
+os.makedirs(dst_syn, exist_ok=True)
+for fn in ("real.csv", "test.csv", "val.csv"):
+    shutil.copy(os.path.join(src, fn), os.path.join(dst_syn, fn))
+os.chdir(rt)
+os.environ["PYTHONPATH"] = rt + os.pathsep + os.environ.get("PYTHONPATH", "")
+os.environ["EFVFM_SMOKE_STEPS"] = "100"
+os.environ["EFVFM_ADAPTER_TRAIN"] = "1"
+os.environ.setdefault("EFVFM_TRAIN_BATCH_SIZE", "64")
+os.environ.setdefault("EFVFM_SAMPLE_BATCH_SIZE", "64")
+os.environ.setdefault("EFVFM_EVAL_NUM_SAMPLES", "512")
+os.environ.setdefault("EFVFM_ODE_FALLBACK", "1")
+os.environ.setdefault("EFVFM_RK4_STEPS", "32")
+subprocess.check_call([
+    sys.executable, os.path.join(rt, "main.py"),
+    "--dataname", name, "--mode", "train", "--gpu", "0",
+    "--no_wandb", "--exp_name", r"adapter_efvfm",
+])
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/gen_20260513_134510.log b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/gen_20260513_134510.log
new file mode 100644
index 0000000000000000000000000000000000000000..d31e2304e37db159ac337337086b2d2c494af0bd
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/gen_20260513_134510.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5d4d9c0a7078479e707c6fe053e98b87988e02a212fc6ec051ea5c681907c18
+size 119152
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/input_snapshot.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..432d53efc030134fb8870cae851b097386429807
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/input_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17cc1ecbae953a319c94a177dada630eed5c295e9c149751ff9ae91971961a4e
+size 1370
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/models_tabbyflow/trained.pt b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/models_tabbyflow/trained.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f063a878c470d9d40f79db79881596e1e497dffd
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/models_tabbyflow/trained.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adb734fec12f2251befd371dca69e481b19464aded2a6823105a01b4be6ddbe5
+size 40
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..319c73410effc766e0bbaedef8dbbbf8451d5f74
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8e13e1c6e622699e221296c60f78826768eb2924cb4fed7c34a19ef5f2b80a4
+size 16264
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/public_gate_report.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..ab1db23ecc6ed1226256cf612bb1c9d692521729
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/public_gate_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea494477cf967f69e54af6c8693533ac8881a2a84c785ee01246213ed0bd2dde
+size 919
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..63c3d5e67b9573266ade7d8ba8f6c9c3d96cb8ec
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/public_gate/staged_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47687546bfbc7d23132b5579f54b24a12cda2925a5a3599965e1934c18cf470e
+size 17100
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/run_config.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/run_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..44929c980405a8c902a3b61184d901b2d67befd6
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/run_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97003a4b29676a8602b5b5f007e1a6594b342f3fadd6c684967d7be3ee371770
+size 2199
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/runtime_result.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..493700ce1224e68af4f39b3a78312265da249057
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/runtime_result.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eeccc2e3a5d2300810d83f47babeb9a479eb64071019f32e5b5ba4256d7330ff
+size 936
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/staged_features.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..f901a56424bda5eddcc5b38ff8bbca0b186bd0ac
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/staged_features.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fed791ef5182375f73ad93592b97cc6ac631c25c28e2e31e2c97e78c3a29855b
+size 2784
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/test.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/train.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/val.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_report.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..5b1b9693acad4c0f20f839221d28118440d0b10d
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d97a3d84c9e226efac6c8be41d5d307f00ff7fd8cbf4d0d86b8ead7a67ea20a
+size 329
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_transforms_applied.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..6f3aaa01a627a955d2022165805d2786c0da78ba
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/adapter_transforms_applied.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/model_input_manifest.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..6de4a5f1624265f4f4015f098ba19a44ba52abbc
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/staged/tabbyflow/model_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f904c74291b72b1536367773152cb2b94ad5443dcc9632e4763e2f71293f8cdf
+size 17305
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow-n16-227845-20260513_134510.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow-n16-227845-20260513_134510.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d86134a1f3eb8547842688f01c16786abd79beac
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow-n16-227845-20260513_134510.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de2b761724b561f959e01aabcc40372fa6e4d059711d96ea485fa903d1fccd0d
+size 75469976
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow_train_meta.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow_train_meta.json
new file mode 100644
index 0000000000000000000000000000000000000000..10be4ea2365e0712c0ae03a43d1d65567b3f678a
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabbyflow_train_meta.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68cd417c0600c91c7a356ca5178122de0a2e27272eeb371a99f9a3ad92c33654
+size 414
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_test.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..456c96321c974010e2a9fa514f0299fad92beb18
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f612f9120f3c6c0b1a9bcf3ea292af3d279674a92e0022faff9a6ae8260ae16e
+size 3417968
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_train.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..88ccc2af99fc1d869424b9585882d50ab8d3c53f
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca0368f23f7b1f2e91278788c5e5586d242dd4c63ecdea703f11014da63c1771
+size 27341528
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_val.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..bf3a294e26e5fd6094fc4807ce1df59135b4c379
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/X_num_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26e7f4c1afa2f9e2db45009d21e961511bce579b9215379827b57a54171c986c
+size 3417728
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/info.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..1a8e13734371c95ae04b9588d0a1bca2e45f7c61
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/info.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd24f6c9bf68c4f74910690803d3750434847954576c6b27ae1989d4c2d08f3c
+size 5350
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/real.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/real.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/real.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/staged_features.json b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..f901a56424bda5eddcc5b38ff8bbca0b186bd0ac
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/staged_features.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fed791ef5182375f73ad93592b97cc6ac631c25c28e2e31e2c97e78c3a29855b
+size 2784
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/test.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/train.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/val.csv b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_test.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..a790b7c3bdf7de5da8cbbb3a2c7dfe1596f0e32e
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e68a11c97bcc5ff28be86a44450abb4663c29c718f1359d168c0b37036e78b78
+size 227984
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_train.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8076c51fb125c43b24274c9061b5993e56d63365
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5320385ca99e9d5210ae94d6b3338d824757df7cb44c55282bf93277d9d3249
+size 1822888
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_val.npy b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..316a8d952b3f622d33c1536df25da7a1ef9bdd92
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/tabular_bundle/pipeline_n16/y_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149a45178d1a5b41d0cc4d0bef92144dcc25e12f312e6e15b753be2dcbce5c23
+size 227968
diff --git a/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/train_20260513_131701.log b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/train_20260513_131701.log
new file mode 100644
index 0000000000000000000000000000000000000000..e1bf93f8ed91b625e3c408a754f20d016d5d34c7
--- /dev/null
+++ b/SynthData0523/main/n16/tabbyflow/tabbyflow-n16-20260513_131635/train_20260513_131701.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:326c3899793d4384a04806250c92acce68b66c729ee6a8bb6430f082a727ed9d
+size 4903124
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_sample.py b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_sample.py
new file mode 100644
index 0000000000000000000000000000000000000000..1dcc7a4116cb41f5191a50ded1fa56448b70c861
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_sample.py
@@ -0,0 +1,66 @@
+import os, sys, subprocess, json
+import numpy as np
+import pandas as pd
+
+tabddpm_root = "/workspace/tabddpm/code"
+assert os.path.isdir(tabddpm_root), f"TabDDPM source not mounted: {tabddpm_root}"
+env = os.environ.copy()
+env["PYTHONPATH"] = tabddpm_root + (os.pathsep + env.get("PYTHONPATH", ""))
+
+# Reuse the compat wrapper (patches collections.Sequence for skorch)
+wrapper = os.path.join(tabddpm_root, "_compat_run.py")
+if not os.path.exists(wrapper):
+    with open(wrapper, "w") as f:
+        f.write(
+            "import collections, collections.abc\n"
+            "for _a in ('Sequence','MutableSequence','MutableMapping','Mapping',"
+            "'MutableSet','Set','Callable','Iterable','Iterator'):\n"
+            "    if not hasattr(collections, _a): setattr(collections, _a, getattr(collections.abc, _a, None))\n"
+            "import sys, runpy\n"
+            "sys.argv = sys.argv[1:]\n"
+            "runpy.run_path(sys.argv[0], run_name='__main__')\n"
+        )
+
+print(f"[TabDDPM] Sampling 227845 rows")
+ret = subprocess.run(
+    [sys.executable, wrapper, "scripts/pipeline.py",
+     "--config", "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_080506.toml",
+     "--sample"],
+    cwd=tabddpm_root,
+    env=env
+)
+if ret.returncode != 0:
+    sys.exit(ret.returncode)
+
+# 将 .npy 输出转为 CSV（npy 在 TabDDPM 的 parent_dir，即 npy_dir）
+info_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data/info.json"
+with open(info_path) as f:
+    info = json.load(f)
+
+output_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+col_names = info.get("column_names", [])
+
+parts = []
+x_num_path = os.path.join(output_dir, "X_num_train.npy")
+x_cat_path = os.path.join(output_dir, "X_cat_train.npy")
+y_path = os.path.join(output_dir, "y_train.npy")
+
+if os.path.exists(x_num_path):
+    parts.append(np.load(x_num_path, allow_pickle=True))
+if os.path.exists(x_cat_path):
+    parts.append(np.load(x_cat_path, allow_pickle=True).astype(float))
+if os.path.exists(y_path):
+    y = np.load(y_path, allow_pickle=True)
+    parts.append(y.reshape(-1, 1) if y.ndim == 1 else y)
+
+if parts:
+    combined = np.concatenate(parts, axis=1)
+    if col_names and len(col_names) == combined.shape[1]:
+        df = pd.DataFrame(combined, columns=col_names)
+    else:
+        df = pd.DataFrame(combined)
+    df.to_csv("/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv", index=False)
+    print(f"[TabDDPM] Saved {len(df)} rows -> /work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv")
+else:
+    print("[TabDDPM] WARNING: No output .npy files found")
+    sys.exit(1)
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_train.py b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..c74b289d4d61db77d43d38b0e990881186f34e56
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/_tabddpm_train.py
@@ -0,0 +1,32 @@
+import os, sys, subprocess
+
+tabddpm_root = "/workspace/tabddpm/code"
+assert os.path.isdir(tabddpm_root), f"TabDDPM source not mounted: {tabddpm_root}"
+env = os.environ.copy()
+env["PYTHONPATH"] = tabddpm_root + (os.pathsep + env.get("PYTHONPATH", ""))
+
+# Write a wrapper that patches collections.Sequence (removed in Python 3.10+)
+# before running pipeline.py - needed because skorch uses old API
+wrapper = os.path.join(tabddpm_root, "_compat_run.py")
+with open(wrapper, "w") as f:
+    f.write(
+        "import collections, collections.abc\n"
+        "for _a in ('Sequence','MutableSequence','MutableMapping','Mapping',"
+        "'MutableSet','Set','Callable','Iterable','Iterator'):\n"
+        "    if not hasattr(collections, _a): setattr(collections, _a, getattr(collections.abc, _a, None))\n"
+        "import sys, runpy\n"
+        "sys.argv = sys.argv[1:]\n"
+        "runpy.run_path(sys.argv[0], run_name='__main__')\n"
+    )
+
+print(f"[TabDDPM] Training, config=/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/config.toml")
+ret = subprocess.run(
+    [sys.executable, wrapper, "scripts/pipeline.py",
+     "--config", "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/config.toml",
+     "--train"],
+    cwd=tabddpm_root,
+    env=env
+)
+if ret.returncode != 0:
+    sys.exit(ret.returncode)
+print("[TabDDPM] Training complete")
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config.toml b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config.toml
new file mode 100644
index 0000000000000000000000000000000000000000..149ce527344edff5a82d775270b6d8d49161cd8d
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config.toml
@@ -0,0 +1,39 @@
+seed = 0
+parent_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+real_data_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data"
+model_type = "mlp"
+num_numerical_features = 30
+device = "cuda:0"
+
+[model_params]
+d_in = 30
+num_classes = 2
+is_y_cond = true
+
+[model_params.rtdl_params]
+d_layers = [256, 256]
+dropout = 0.0
+
+[diffusion_params]
+num_timesteps = 1000
+gaussian_loss_type = "mse"
+
+[train.main]
+steps = 5000
+lr = 0.001
+weight_decay = 0.0
+batch_size = 256
+
+[train.T]
+seed = 0
+normalization = "quantile"
+num_nan_policy = "__none__"
+cat_nan_policy = "__none__"
+cat_min_frequency = "__none__"
+cat_encoding = "__none__"
+y_policy = "default"
+
+[sample]
+num_samples = 1000
+batch_size = 1000
+seed = 0
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260424_212203.toml b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260424_212203.toml
new file mode 100644
index 0000000000000000000000000000000000000000..8b161a54dd94fda70c21bbe875ebab2f29f3627c
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260424_212203.toml
@@ -0,0 +1,39 @@
+seed = 0
+parent_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+real_data_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data"
+model_type = "mlp"
+num_numerical_features = 30
+device = "cuda:0"
+
+[model_params]
+d_in = 30
+num_classes = 2
+is_y_cond = true
+
+[model_params.rtdl_params]
+d_layers = [256, 256]
+dropout = 0.0
+
+[diffusion_params]
+num_timesteps = 1000
+gaussian_loss_type = "mse"
+
+[train.main]
+steps = 5000
+lr = 0.001
+weight_decay = 0.0
+batch_size = 256
+
+[train.T]
+seed = 0
+normalization = "quantile"
+num_nan_policy = "__none__"
+cat_nan_policy = "__none__"
+cat_min_frequency = "__none__"
+cat_encoding = "__none__"
+y_policy = "default"
+
+[sample]
+num_samples = 227845
+batch_size = 1000
+seed = 0
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_033728.toml b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_033728.toml
new file mode 100644
index 0000000000000000000000000000000000000000..8b161a54dd94fda70c21bbe875ebab2f29f3627c
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_033728.toml
@@ -0,0 +1,39 @@
+seed = 0
+parent_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+real_data_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data"
+model_type = "mlp"
+num_numerical_features = 30
+device = "cuda:0"
+
+[model_params]
+d_in = 30
+num_classes = 2
+is_y_cond = true
+
+[model_params.rtdl_params]
+d_layers = [256, 256]
+dropout = 0.0
+
+[diffusion_params]
+num_timesteps = 1000
+gaussian_loss_type = "mse"
+
+[train.main]
+steps = 5000
+lr = 0.001
+weight_decay = 0.0
+batch_size = 256
+
+[train.T]
+seed = 0
+normalization = "quantile"
+num_nan_policy = "__none__"
+cat_nan_policy = "__none__"
+cat_min_frequency = "__none__"
+cat_encoding = "__none__"
+y_policy = "default"
+
+[sample]
+num_samples = 227845
+batch_size = 1000
+seed = 0
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_080506.toml b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_080506.toml
new file mode 100644
index 0000000000000000000000000000000000000000..8b161a54dd94fda70c21bbe875ebab2f29f3627c
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/config_sample_20260425_080506.toml
@@ -0,0 +1,39 @@
+seed = 0
+parent_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+real_data_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data"
+model_type = "mlp"
+num_numerical_features = 30
+device = "cuda:0"
+
+[model_params]
+d_in = 30
+num_classes = 2
+is_y_cond = true
+
+[model_params.rtdl_params]
+d_layers = [256, 256]
+dropout = 0.0
+
+[diffusion_params]
+num_timesteps = 1000
+gaussian_loss_type = "mse"
+
+[train.main]
+steps = 5000
+lr = 0.001
+weight_decay = 0.0
+batch_size = 256
+
+[train.T]
+seed = 0
+normalization = "quantile"
+num_nan_policy = "__none__"
+cat_nan_policy = "__none__"
+cat_min_frequency = "__none__"
+cat_encoding = "__none__"
+y_policy = "default"
+
+[sample]
+num_samples = 227845
+batch_size = 1000
+seed = 0
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_test.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..0c97d692014ec3d4e179ce9c652746c228611d3e
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2188431173531bc74c4341d6379a476cc310bbb40c19080dcdd4e9c3d509804a
+size 3417968
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_train.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_val.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..0e94d7e2e7f98d2dc0af00336f5d8d35f5cf0618
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/X_num_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af8abe13ddd7cf5750c852cca65018967fdda842775879a1e7cad2d4daf9f924
+size 3417728
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/info.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..7191b265bbfc79d4e087881ff5d65d79e7fe6c01
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/info.json
@@ -0,0 +1,77 @@
+{
+  "name": "benchmark_dataset",
+  "task_type": "multiclass",
+  "n_num_features": 30,
+  "n_cat_features": 0,
+  "train_size": 227845,
+  "num_col_idx": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29
+  ],
+  "cat_col_idx": [],
+  "target_col_idx": [
+    30
+  ],
+  "column_names": [
+    "Time",
+    "V1",
+    "V2",
+    "V3",
+    "V4",
+    "V5",
+    "V6",
+    "V7",
+    "V8",
+    "V9",
+    "V10",
+    "V11",
+    "V12",
+    "V13",
+    "V14",
+    "V15",
+    "V16",
+    "V17",
+    "V18",
+    "V19",
+    "V20",
+    "V21",
+    "V22",
+    "V23",
+    "V24",
+    "V25",
+    "V26",
+    "V27",
+    "V28",
+    "Amount",
+    "Class"
+  ],
+  "num_classes": 2
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_test.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..a790b7c3bdf7de5da8cbbb3a2c7dfe1596f0e32e
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e68a11c97bcc5ff28be86a44450abb4663c29c718f1359d168c0b37036e78b78
+size 227984
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_train.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8076c51fb125c43b24274c9061b5993e56d63365
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5320385ca99e9d5210ae94d6b3338d824757df7cb44c55282bf93277d9d3249
+size 1822888
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_val.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..316a8d952b3f622d33c1536df25da7a1ef9bdd92
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/data/y_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149a45178d1a5b41d0cc4d0bef92144dcc25e12f312e6e15b753be2dcbce5c23
+size 227968
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260321_164225.log b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260321_164225.log
new file mode 100644
index 0000000000000000000000000000000000000000..13ffc2c53555763d3965d6e9cbf167f256499533
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260321_164225.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4863abd5a9d20b92b8799570d195a9a67afa512f98cd213d776cc1ba2d99571e
+size 21560
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260424_212203.log b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260424_212203.log
new file mode 100644
index 0000000000000000000000000000000000000000..b3fe18dc2b6fb78071a6ffe286676c175db3209a
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260424_212203.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc37ea2374fed11fb6f48523f0f6ff4bb07e9b86cf82347c4a63efac4e6901d3
+size 4788463
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_033728.log b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_033728.log
new file mode 100644
index 0000000000000000000000000000000000000000..158cd0d3c83f5ee27465029ee901d1bb0b2f88f6
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_033728.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85950cf1ab526fc4040e23c1b287d6874442ed77c68faa55d2a3f871eacea172
+size 4788146
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_080506.log b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_080506.log
new file mode 100644
index 0000000000000000000000000000000000000000..6a4aa5186e6f7aa25a762c54bb5d81bd256ed51d
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/gen_20260425_080506.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4d35a3b1e5ab7e5462dcc80ca9883209334924a7c4f505f798cfa7455ddb299
+size 4788241
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/input_snapshot.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..f6e687bbcab32fb551343287ca3cda76d78ff3a4
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/input_snapshot.json
@@ -0,0 +1,36 @@
+{
+  "dataset_id": "n16",
+  "model": "tabddpm",
+  "inputs": {
+    "train_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+      "exists": true,
+      "size": 120434969,
+      "sha256": "d33c0f51e8e82d110fc3787d448c55eb646bb006d4142775762e29d38619dab7"
+    },
+    "val_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+      "exists": true,
+      "size": 15054728,
+      "sha256": "ba8c9f03e07e66776ad749730d985bb6491b848595a0727e851448fc8c0a28ce"
+    },
+    "test_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv",
+      "exists": true,
+      "size": 15054431,
+      "sha256": "23f5d0e05e2e308538143a51e1ceb6f019680d1e92ef6c37626cc7774f22b45b"
+    },
+    "profile_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_profile.json",
+      "exists": true,
+      "size": 13450,
+      "sha256": "dc84ba22eb87a13dee40814eb0af363bbe651e48726c8f07b641f4e30f33a235"
+    },
+    "contract_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_contract_v1.json",
+      "exists": true,
+      "size": 16859,
+      "sha256": "3e28017c8292679d4e8b4eeca7a3c31c5496db891c09e8b15d30f57e2884b12e"
+    }
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_train.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..c038d10a69923c21fadd9223059f74ab59a178eb
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8909e35f37abb454809ec5f00d53d8badf8e535028d18d9d034b8211e1efbe30
+size 54682928
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_unnorm.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_unnorm.npy
new file mode 100644
index 0000000000000000000000000000000000000000..30671debae6d723fb55f207a7a5bd33e71350620
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/X_num_unnorm.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7af4a07ba41cbe23517c664e5a5c97ee1149cc6b6178101ebc7427914ec1df73
+size 54682928
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/config.toml b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/config.toml
new file mode 100644
index 0000000000000000000000000000000000000000..8b161a54dd94fda70c21bbe875ebab2f29f3627c
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/config.toml
@@ -0,0 +1,39 @@
+seed = 0
+parent_dir = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/output"
+real_data_path = "/work/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/data"
+model_type = "mlp"
+num_numerical_features = 30
+device = "cuda:0"
+
+[model_params]
+d_in = 30
+num_classes = 2
+is_y_cond = true
+
+[model_params.rtdl_params]
+d_layers = [256, 256]
+dropout = 0.0
+
+[diffusion_params]
+num_timesteps = 1000
+gaussian_loss_type = "mse"
+
+[train.main]
+steps = 5000
+lr = 0.001
+weight_decay = 0.0
+batch_size = 256
+
+[train.T]
+seed = 0
+normalization = "quantile"
+num_nan_policy = "__none__"
+cat_nan_policy = "__none__"
+cat_min_frequency = "__none__"
+cat_encoding = "__none__"
+y_policy = "default"
+
+[sample]
+num_samples = 227845
+batch_size = 1000
+seed = 0
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/data/info.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/data/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..7191b265bbfc79d4e087881ff5d65d79e7fe6c01
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/data/info.json
@@ -0,0 +1,77 @@
+{
+  "name": "benchmark_dataset",
+  "task_type": "multiclass",
+  "n_num_features": 30,
+  "n_cat_features": 0,
+  "train_size": 227845,
+  "num_col_idx": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29
+  ],
+  "cat_col_idx": [],
+  "target_col_idx": [
+    30
+  ],
+  "column_names": [
+    "Time",
+    "V1",
+    "V2",
+    "V3",
+    "V4",
+    "V5",
+    "V6",
+    "V7",
+    "V8",
+    "V9",
+    "V10",
+    "V11",
+    "V12",
+    "V13",
+    "V14",
+    "V15",
+    "V16",
+    "V17",
+    "V18",
+    "V19",
+    "V20",
+    "V21",
+    "V22",
+    "V23",
+    "V24",
+    "V25",
+    "V26",
+    "V27",
+    "V28",
+    "Amount",
+    "Class"
+  ],
+  "num_classes": 2
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/info.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..7191b265bbfc79d4e087881ff5d65d79e7fe6c01
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/info.json
@@ -0,0 +1,77 @@
+{
+  "name": "benchmark_dataset",
+  "task_type": "multiclass",
+  "n_num_features": 30,
+  "n_cat_features": 0,
+  "train_size": 227845,
+  "num_col_idx": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29
+  ],
+  "cat_col_idx": [],
+  "target_col_idx": [
+    30
+  ],
+  "column_names": [
+    "Time",
+    "V1",
+    "V2",
+    "V3",
+    "V4",
+    "V5",
+    "V6",
+    "V7",
+    "V8",
+    "V9",
+    "V10",
+    "V11",
+    "V12",
+    "V13",
+    "V14",
+    "V15",
+    "V16",
+    "V17",
+    "V18",
+    "V19",
+    "V20",
+    "V21",
+    "V22",
+    "V23",
+    "V24",
+    "V25",
+    "V26",
+    "V27",
+    "V28",
+    "Amount",
+    "Class"
+  ],
+  "num_classes": 2
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/loss.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/loss.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e80c801b53680e3af06e57a093ab6a8167ae5cc0
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/loss.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4166639e5649f3958c3afa5919ef942c2753c3cfa3149e8d569b8aa18505ae10
+size 1255
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model.pt b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5cf43491e5309eb2b3a57c0c0680598e08af994a
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83d4c7863b137e193f1c850f4ccbc368603372302df7b7fe56eac6cc667db364
+size 579734
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model_ema.pt b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model_ema.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c411ea8222062affd658156bc6bc04ae49ca5e9e
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/model_ema.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fe039a1282633c87aaf617d2c1d1fb03d9435343b5803c6cae011d31bd96907
+size 580578
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/y_train.npy b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/y_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..d802594dbe34fd06e65bfdb27fbd320c183e1a9f
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/output/y_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6f8c62389b4d576ac179d1d3d05554b949134ff391a95af60f8cc9c69af1641
+size 1822888
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..d391c1ee1422eb5ee4ce05e4391604265338d18a
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,655 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "columns": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/public_gate_report.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..77d1958fecefe23b3791fd07b5a6eaa8296b1d05
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/public_gate_report.json
@@ -0,0 +1,37 @@
+{
+  "dataset_id": "n16",
+  "status": "pass",
+  "checks": [
+    {
+      "check_id": "PG001_csv_parse_ok",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG002_split_header_consistent",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG003_profile_header_match",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG004_missing_token_normalized",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG005_semantic_type_validated",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG006_target_defined_and_valid",
+      "status": "pass"
+    }
+  ],
+  "target_column": "Class",
+  "task_type": "classification",
+  "input_splits": {
+    "train": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+    "val": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+    "test": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..a5eddb18e499c0f11277fee2fa55ec16f8b168f6
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/staged_input_manifest.json
@@ -0,0 +1,660 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/public_gate_report.json",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/runtime_result.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..d48f6d810578592b7eb7265fc5ae5d07e2081caf
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/runtime_result.json
@@ -0,0 +1,14 @@
+{
+  "dataset_id": "n16",
+  "model": "tabddpm",
+  "run_id": "tabddpm-n16-20260321_163935",
+  "public_gate_status": "pass",
+  "adapter_ready_status": "pass",
+  "train_status": "skipped",
+  "generate_status": "success",
+  "reason_code": null,
+  "reason_detail": null,
+  "artifacts": {
+    "synthetic_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/staged_features.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..4fa94dfbf395e3c2359e2e16c0488f3f0b698f00
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/staged_features.json
@@ -0,0 +1,157 @@
+[
+  {
+    "feature_name": "Time",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V1",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V2",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V3",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V4",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V5",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V6",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V7",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V8",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V9",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V10",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V11",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V12",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V13",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V14",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V15",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V16",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V17",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V18",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V19",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V20",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V21",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V22",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V23",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V24",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V25",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V26",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V27",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V28",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Amount",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Class",
+    "data_type": "binary",
+    "is_target": true
+  }
+]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_report.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..61cf3580859bcd59d33a96d72f1760ffd4a8d6e1
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_report.json
@@ -0,0 +1,7 @@
+{
+  "adapter_ready_status": "pass",
+  "adapter_fail_reason_code": null,
+  "adapter_fail_detail": null,
+  "adapter_transforms_applied": [],
+  "model_input_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/model_input_manifest.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_transforms_applied.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..0637a088a01e8ddab3bf3fa98dbe804cbde1a0dc
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/adapter_transforms_applied.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/model_input_manifest.json b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..19b5ecece1173584324f6f065dbcffd26a9d26f6
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/staged/tabddpm/model_input_manifest.json
@@ -0,0 +1,662 @@
+{
+  "dataset_id": "n16",
+  "model": "tabddpm",
+  "target_column": "Class",
+  "task_type": "classification",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ],
+  "public_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/staged_input_manifest.json",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabddpm/tabddpm-n16-20260321_163935/public_gate/public_gate_report.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-1000-20260321_164225.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-1000-20260321_164225.csv
new file mode 100644
index 0000000000000000000000000000000000000000..a16f787422f280abaa0714a7dd34976b7184be30
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-1000-20260321_164225.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c1a918cb910db02ab9939c0f89073dc522c0365408ac82a915d750f09b251ff
+size 587026
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv
new file mode 100644
index 0000000000000000000000000000000000000000..27d4a9b6c157abc374bc95520e6131e8ac615f87
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/tabddpm-n16-227845-20260425_080506.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d545870409ae2126d6d2215f238d757ce7ec5e85e63c247dc11c6af22d564521
+size 130288898
diff --git a/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/train_20260321_164024.log b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/train_20260321_164024.log
new file mode 100644
index 0000000000000000000000000000000000000000..491068e265cf2145c073ca63e05d7411347534d1
--- /dev/null
+++ b/SynthData0523/main/n16/tabddpm/tabddpm-n16-20260321_163935/train_20260321_164024.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d97f19f5263cc42094902e2af95a80c4690e0fd6154dc21bf0fcfeb7199dec3
+size 1124
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_gen.py b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_gen.py
new file mode 100644
index 0000000000000000000000000000000000000000..a129ce604a2b25edfb0f99038ebd8b4d0e4f232b
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_gen.py
@@ -0,0 +1,36 @@
+
+import os, shutil, subprocess, sys
+td = r"/workspace/TabDiff"
+name = r"pipeline_n16"
+src = r"/work/output-Benchmark-trainonly-v1/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16"
+dst_data = os.path.join(td, "data", name)
+dst_syn = os.path.join(td, "synthetic", name)
+shutil.rmtree(dst_data, ignore_errors=True)
+shutil.copytree(src, dst_data)
+os.makedirs(dst_syn, exist_ok=True)
+for fn in ("real.csv", "test.csv", "val.csv"):
+    shutil.copy(os.path.join(src, fn), os.path.join(dst_syn, fn))
+os.chdir(td)
+os.environ["PYTHONPATH"] = td + os.pathsep + os.environ.get("PYTHONPATH", "")
+subprocess.check_call([
+    sys.executable, "-m", "tabdiff.main",
+    "--dataname", name, "--mode", "test", "--gpu", "0",
+    "--no_wandb", "--exp_name", r"adapter_learnable",
+    "--ckpt_path", r"/workspace/TabDiff/tabdiff/ckpt/pipeline_n16/adapter_learnable/model_500.pt",
+    "--num_samples_to_generate", str(int(227845)),
+])
+# test() 写入 tabdiff/result/<dataname>/<exp>/<epoch>/samples.csv
+import glob as g
+base = os.path.join(td, "tabdiff", "result", name, r"adapter_learnable")
+best = None
+best_t = -1.0
+for root, _, files in os.walk(base):
+    if "samples.csv" in files:
+        p = os.path.join(root, "samples.csv")
+        t = os.path.getmtime(p)
+        if t > best_t:
+            best_t = t
+            best = p
+if not best:
+    raise SystemExit("tabdiff: no samples.csv under " + base)
+shutil.copy(best, r"/work/output-Benchmark-trainonly-v1/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff-n16-227845-20260501_194646.csv")
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_train.py b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..788983fd8eb7f70622bd1b406763ee537f5e59f9
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/_tabdiff_train.py
@@ -0,0 +1,21 @@
+
+import os, shutil, subprocess, sys
+td = r"/workspace/TabDiff"
+name = r"pipeline_n16"
+src = r"/work/output-Benchmark-trainonly-v1/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16"
+dst_data = os.path.join(td, "data", name)
+dst_syn = os.path.join(td, "synthetic", name)
+shutil.rmtree(dst_data, ignore_errors=True)
+shutil.copytree(src, dst_data)
+os.makedirs(dst_syn, exist_ok=True)
+for fn in ("real.csv", "test.csv", "val.csv"):
+    shutil.copy(os.path.join(src, fn), os.path.join(dst_syn, fn))
+os.chdir(td)
+os.environ["PYTHONPATH"] = td + os.pathsep + os.environ.get("PYTHONPATH", "")
+os.environ["TABDIFF_SMOKE_STEPS"] = "500"
+os.environ["TABDIFF_ADAPTER_TRAIN"] = "1"
+subprocess.check_call([
+    sys.executable, "-m", "tabdiff.main",
+    "--dataname", name, "--mode", "train", "--gpu", "0",
+    "--no_wandb", "--exp_name", r"adapter_learnable",
+])
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/gen_20260501_194646.log b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/gen_20260501_194646.log
new file mode 100644
index 0000000000000000000000000000000000000000..ac3103058046f71f48099dbef8fc666bdb801fbf
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/gen_20260501_194646.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f27e10ed7d4d73a44bf8fa55b6dd4151d827d32c060f9fb3ee03173df8f2502
+size 23334
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/input_snapshot.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..05d301803c3467b7b72ea47f9ccfd417b65ae5ff
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/input_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f53c08364931bbb099f7509bea1f15fb88b1feadbd1b228f363c550a8266a3e
+size 1368
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/models_tabdiff/trained.pt b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/models_tabdiff/trained.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bc319a55a9c7152a1137f968b6fa1c735125ad11
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/models_tabdiff/trained.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:306660b8aad4549267c68390b017e15ddb9bc06a02c80515eb72a97ae31a81eb
+size 74
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..319c73410effc766e0bbaedef8dbbbf8451d5f74
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8e13e1c6e622699e221296c60f78826768eb2924cb4fed7c34a19ef5f2b80a4
+size 16264
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/public_gate_report.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..ab1db23ecc6ed1226256cf612bb1c9d692521729
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/public_gate_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea494477cf967f69e54af6c8693533ac8881a2a84c785ee01246213ed0bd2dde
+size 919
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..a2f824389130c314c86d414442362d802f6014db
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/public_gate/staged_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b8d34f97809f6eab3626ece2db0e8664fd19961c9b8272a392b5d9bda87fb3e
+size 17080
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/runtime_result.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..79b51377f57397d699e0569cdd107936e5a251fd
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/runtime_result.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d2a4f5ffe611b6d193bcac2605f75cba99baa508f71ca720343c22180508228
+size 918
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/staged_features.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..f901a56424bda5eddcc5b38ff8bbca0b186bd0ac
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/staged_features.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fed791ef5182375f73ad93592b97cc6ac631c25c28e2e31e2c97e78c3a29855b
+size 2784
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/test.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/train.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/val.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_report.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..1cc38a798e5220a799a2bb16ad70665b2978985b
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71ff7498231d5cfabe3825448377a8ca9945cf5e0311249cda2bc1a67cf27745
+size 323
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_transforms_applied.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..6f3aaa01a627a955d2022165805d2786c0da78ba
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/adapter_transforms_applied.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/model_input_manifest.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..fab98f0926c394e08fff7e4e03cb696e34ec5471
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/staged/tabdiff/model_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f98509cbc0b7e0d723b06119b484ced4692b6b87097bacfeee7c7c45b1fe493b
+size 17279
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff-n16-227845-20260501_194646.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff-n16-227845-20260501_194646.csv
new file mode 100644
index 0000000000000000000000000000000000000000..902490e6611d80a2f797e34f95c0acb4c5a6bfb2
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff-n16-227845-20260501_194646.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37bf3b8801fb6f3d1816a1e46d63ef92e850479e64c1668f7a2ba28619683e9c
+size 75719995
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff_train_meta.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff_train_meta.json
new file mode 100644
index 0000000000000000000000000000000000000000..c5d59c0e0da4199e60602a3f7f8732a031a50103
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabdiff_train_meta.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111109bb4fe3b3940b95f35327d76ad8c37bf510ac4473706df75ec5daaceb6a
+size 83
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_test.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..c8dfeec71b09916c114b51f4f45836194a1cfa29
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d584e904bdf7d7d9317093dc13374950e0bc9cf30059bf8c421fc3a52469f2d
+size 128
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_train.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..c8dfeec71b09916c114b51f4f45836194a1cfa29
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d584e904bdf7d7d9317093dc13374950e0bc9cf30059bf8c421fc3a52469f2d
+size 128
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_val.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..c8dfeec71b09916c114b51f4f45836194a1cfa29
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_cat_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d584e904bdf7d7d9317093dc13374950e0bc9cf30059bf8c421fc3a52469f2d
+size 128
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_test.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_train.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_val.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/X_num_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/info.json b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..39b739f78a9a4249dad30b5abc48c01ac35266b2
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/info.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f6a54089ff2eca532f383f3abf6f1d2cc0900ff5bce192a91f34d14967c8778
+size 5472
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/real.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/real.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9e968920d10cb9250607660b5b73c1374ee66822
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/real.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6d729ed427e4c13d1bef7313c96b10abb9076e8ad66a549d084db80859f8f01
+size 120653540
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/test.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9e968920d10cb9250607660b5b73c1374ee66822
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6d729ed427e4c13d1bef7313c96b10abb9076e8ad66a549d084db80859f8f01
+size 120653540
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/val.csv b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..9e968920d10cb9250607660b5b73c1374ee66822
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6d729ed427e4c13d1bef7313c96b10abb9076e8ad66a549d084db80859f8f01
+size 120653540
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_test.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8076c51fb125c43b24274c9061b5993e56d63365
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5320385ca99e9d5210ae94d6b3338d824757df7cb44c55282bf93277d9d3249
+size 1822888
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_train.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8076c51fb125c43b24274c9061b5993e56d63365
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5320385ca99e9d5210ae94d6b3338d824757df7cb44c55282bf93277d9d3249
+size 1822888
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_val.npy b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_val.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8076c51fb125c43b24274c9061b5993e56d63365
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/tabular_bundle/pipeline_n16/y_val.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5320385ca99e9d5210ae94d6b3338d824757df7cb44c55282bf93277d9d3249
+size 1822888
diff --git a/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/train_20260501_191724.log b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/train_20260501_191724.log
new file mode 100644
index 0000000000000000000000000000000000000000..6c1084eac1fd79624f66491c19b5b0e8408259f2
--- /dev/null
+++ b/SynthData0523/main/n16/tabdiff/tabdiff-n16-20260501_191651/train_20260501_191724.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12d7be2bf78d652d6cff2e43785fc884d21968a6ba28348223d08c5b928754b9
+size 6258086
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/_tabpfgen_generate.py b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/_tabpfgen_generate.py
new file mode 100644
index 0000000000000000000000000000000000000000..a3d9a8502ec9fbe6b752c1a2a87700e7a97d97ab
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/_tabpfgen_generate.py
@@ -0,0 +1,131 @@
+import os
+import numpy as np
+import pandas as pd
+import json
+from tabpfgen import TabPFGen
+
+df = pd.read_csv("/work/output-Benchmark-trainonly-v1/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/train.csv")
+target_col = "Class"
+
+target_missing = df[target_col].isna()
+if target_missing.any():
+    dropped = int(target_missing.sum())
+    df = df.loc[~target_missing].copy()
+    print(
+        f"[TabPFGen] Dropped {dropped} rows with missing target '{target_col}'"
+    )
+if df.empty:
+    raise ValueError(
+        f"[TabPFGen] No rows remain after dropping missing target '{target_col}'"
+    )
+
+feature_cols = [c for c in df.columns if c != target_col]
+
+cat_encodings = {}
+for col in feature_cols:
+    if df[col].dtype == object or str(df[col].dtype) == 'category':
+        cats = sorted(df[col].dropna().unique().tolist(), key=str)
+        cat_map = {v: i for i, v in enumerate(cats)}
+        df[col] = df[col].map(cat_map).astype(float)
+        cat_encodings[col] = cats
+        print(f"[TabPFGen] Label-encoded '{col}' ({len(cats)} categories)")
+
+target_cats = None
+if df[target_col].dtype == object or str(df[target_col].dtype) == 'category':
+    cats = sorted(df[target_col].dropna().unique().tolist(), key=str)
+    t_map = {v: i for i, v in enumerate(cats)}
+    df[target_col] = df[target_col].map(t_map).astype(float)
+    target_cats = cats
+    print(f"[TabPFGen] Label-encoded target '{target_col}' ({len(cats)} categories)")
+
+X = df[feature_cols].values.astype(np.float32)
+y = df[target_col].values
+fit_rows_cap = max(1, int(os.environ.get("TABPFGEN_FIT_MAX_ROWS", "50000")))
+if len(X) > fit_rows_cap:
+    rng = np.random.default_rng(42)
+    idx = np.sort(rng.choice(len(X), size=fit_rows_cap, replace=False))
+    X = X[idx]
+    y = y[idx]
+    print(f"[TabPFGen] Downsampled fit rows -> {len(X)} (cap={fit_rows_cap})")
+target_n = int(227845)
+
+for i in range(X.shape[1]):
+    col_vals = X[:, i]
+    mask = np.isnan(col_vals)
+    if mask.any():
+        mean_val = np.nanmean(col_vals)
+        X[mask, i] = mean_val if not np.isnan(mean_val) else 0.0
+
+chunk_rows = max(1, int(os.environ.get("TABPFGEN_GEN_CHUNK_ROWS", "256")))
+device = (os.environ.get("TABPFGEN_DEVICE") or "auto").strip() or "auto"
+
+n_sgld_steps = max(1, int(os.environ.get("TABPFGEN_N_SGLD_STEPS", "1000")))
+sgld_step_size = float(os.environ.get("TABPFGEN_SGLD_STEP_SIZE", "0.01"))
+sgld_noise_scale = float(os.environ.get("TABPFGEN_SGLD_NOISE_SCALE", "0.01"))
+
+# TabPFGen v0.1.x API：仅支持 n_sgld_steps / sgld_* / device。
+# （旧版脚本中的 energy_*_chunk 与上游 TabPFGen 不一致，会导致 TypeError。）
+gen = TabPFGen(
+    n_sgld_steps=n_sgld_steps,
+    sgld_step_size=sgld_step_size,
+    sgld_noise_scale=sgld_noise_scale,
+    device=device,
+)
+
+print(
+    f"[TabPFGen] Generating {target_n} rows via generate_classification "
+    f"(chunk_rows={chunk_rows}, device={device}, "
+    f"n_sgld_steps={n_sgld_steps}, sgld_step_size={sgld_step_size}, "
+    f"sgld_noise_scale={sgld_noise_scale})"
+)
+x_parts = []
+y_parts = []
+remaining = target_n
+while remaining > 0:
+    take = min(chunk_rows, remaining)
+    X_part, y_part = gen.generate_classification(X, y, n_samples=take)
+    x_parts.append(np.asarray(X_part))
+    y_parts.append(np.asarray(y_part))
+    remaining -= take
+    print(f"[TabPFGen] chunk done: take={take}, remaining={remaining}")
+
+X_syn = np.concatenate(x_parts, axis=0)
+y_syn = np.concatenate(y_parts, axis=0)
+
+syn_df = pd.DataFrame(X_syn, columns=feature_cols)
+syn_df[target_col] = y_syn
+
+for col, cats in cat_encodings.items():
+    codes = np.round(syn_df[col].values).astype(int)
+    codes = np.clip(codes, 0, len(cats) - 1)
+    syn_df[col] = [cats[c] for c in codes]
+
+if target_cats is not None:
+    codes = np.round(syn_df[target_col].values).astype(int)
+    codes = np.clip(codes, 0, len(target_cats) - 1)
+    syn_df[target_col] = [target_cats[c] for c in codes]
+
+if len(syn_df) > target_n:
+    print(f"[TabPFGen] Trimming rows: {len(syn_df)} -> {target_n}")
+    syn_df = syn_df.iloc[:target_n].copy()
+elif len(syn_df) < target_n:
+    deficit = target_n - len(syn_df)
+    print(f"[TabPFGen] Padding rows: {len(syn_df)} -> {target_n} (deficit={deficit})")
+    if len(syn_df) > 0:
+        extra = syn_df.sample(n=deficit, replace=True, random_state=42)
+        syn_df = pd.concat(
+            [syn_df.reset_index(drop=True), extra.reset_index(drop=True)],
+            ignore_index=True,
+        )
+    else:
+        syn_df = df[feature_cols + [target_col]].sample(
+            n=target_n, replace=True, random_state=42
+        ).reset_index(drop=True)
+
+syn_df = syn_df[list(df.columns)]
+if len(syn_df) != target_n:
+    raise RuntimeError(
+        f"[TabPFGen] Row alignment failed: got {len(syn_df)}, expected {target_n}"
+    )
+syn_df.to_csv("/work/output-Benchmark-trainonly-v1/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv", index=False)
+print(f"[TabPFGen] Saved {len(syn_df)} rows -> /work/output-Benchmark-trainonly-v1/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv")
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/gen_20260512_082701.log b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/gen_20260512_082701.log
new file mode 100644
index 0000000000000000000000000000000000000000..6c030ae50e27bddcd202a8cc6829ded724b83d3d
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/gen_20260512_082701.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a9b955727de88a905b1c5414bef6cbc14ba07028fc10cac43f79ee41504f2c7
+size 167899
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/input_snapshot.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..31331a6dfb43036351b4087d889db35a3aeac1fc
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/input_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9440ad7673e96535e30200c330f270f4808a73ac9ffb925228621984cb538e32
+size 1369
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..319c73410effc766e0bbaedef8dbbbf8451d5f74
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8e13e1c6e622699e221296c60f78826768eb2924cb4fed7c34a19ef5f2b80a4
+size 16264
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/public_gate_report.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..ab1db23ecc6ed1226256cf612bb1c9d692521729
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/public_gate_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea494477cf967f69e54af6c8693533ac8881a2a84c785ee01246213ed0bd2dde
+size 919
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..06a82812784171e6df69bd0d8ec82ce37f1a82dc
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/public_gate/staged_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f39285117b27c71635eaefbaaae2be175e33f0dd850f113a63831df95a9c0d
+size 17090
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/run_config.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/run_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..afb49c70e5e07f93d475e87b45b0600d891c8096
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/run_config.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6a833b5f03ae43456581087154b7c41ab8dc3bb506fa4e69aa53dc301c8ea35
+size 2035
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/runtime_result.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..98e3b79253f4b09255da6c30e7b87e11d7fe821b
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/runtime_result.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01ff7ff630cc1122a4bebb525f23f4c86d4c4fd94a3ba5be77f29cd2a701d421
+size 898
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/staged_features.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..f901a56424bda5eddcc5b38ff8bbca0b186bd0ac
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/staged_features.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fed791ef5182375f73ad93592b97cc6ac631c25c28e2e31e2c97e78c3a29855b
+size 2784
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/test.csv b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/train.csv b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/val.csv b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_report.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..3305eaf5155aead2c7690dac8bf5783fd8db70e3
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_report.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38a11c7cf08d03d0ec85c08992484593fdae3dc4c3477f96b0a6133a7394f9ee
+size 326
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_transforms_applied.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..6f3aaa01a627a955d2022165805d2786c0da78ba
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/adapter_transforms_applied.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/model_input_manifest.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..5c93a62a2ff13b01ca53330c95ca91bdf3ec95e7
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/staged/tabpfgen/model_input_manifest.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:374f05ec969fef3853bed5aee327e8da6eba3bdebb2013d76eeb064035f85845
+size 17292
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv
new file mode 100644
index 0000000000000000000000000000000000000000..7f892fb9f050d90418b34fb2c9d3c0e974396d72
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen-n16-227845-20260512_082701.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b491be2c67ea762e6611ea29495a58e2de75d5be4c1cf6f5ffaab7dae21fa273
+size 74362791
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen_meta.json b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen_meta.json
new file mode 100644
index 0000000000000000000000000000000000000000..21bdfa157674bbd23f192422771bdfeadfb79720
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/tabpfgen_meta.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3808608a62abbc822653b44aec906de8036fb08ea6096ac28804fe179ae73030
+size 451
diff --git a/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/train_20260512_082701.log b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/train_20260512_082701.log
new file mode 100644
index 0000000000000000000000000000000000000000..36a33f06a35bfeda863540dfe1734b7b6a8de22f
--- /dev/null
+++ b/SynthData0523/main/n16/tabpfgen/tabpfgen-n16-20260512_082648/train_20260512_082701.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d527895d2f526f8c261c0e64cd5b6ae4ad2810a478e7536eb49f7d1cabe7eea6
+size 599
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_sample.py b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_sample.py
new file mode 100644
index 0000000000000000000000000000000000000000..510c4ac08b77792525813eefe3feafea28ffc728
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_sample.py
@@ -0,0 +1,39 @@
+import os, sys, subprocess
+
+work_dir = "/work/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916"
+dataname = "tabsyn_n16"
+output_csv = "/work/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv"
+tabsyn_root = "/workspace/tabsyn"
+
+assert os.path.exists(tabsyn_root), f"TabSyn source not mounted: {tabsyn_root}"
+
+old = os.environ.get("PYTHONPATH", "")
+os.environ["PYTHONPATH"] = tabsyn_root + (os.pathsep + old if old else "")
+sys.path.insert(0, tabsyn_root)
+
+os.chdir(tabsyn_root)
+
+# Ensure data symlink exists
+data_link = os.path.join(tabsyn_root, "data", dataname)
+data_src = os.path.join(work_dir, "data", dataname)
+os.makedirs(os.path.join(tabsyn_root, "data"), exist_ok=True)
+if os.path.exists(data_link):
+    os.remove(data_link)
+os.symlink(data_src, data_link)
+
+print(f"[TabSyn] Sampling 227845 rows")
+env = os.environ.copy()
+env.setdefault("TABSYN_RESUME", "1")
+ret = subprocess.run(
+    [sys.executable, "main.py",
+     "--dataname", dataname,
+     "--mode", "sample",
+     "--method", "tabsyn",
+     "--gpu", "0",
+     "--save_path", output_csv],
+    cwd=tabsyn_root,
+    env=env
+)
+if ret.returncode != 0:
+    sys.exit(ret.returncode)
+print(f"[TabSyn] Saved -> {output_csv}")
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_train.py b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..1c8bdb43456d42a6bf742f39c9b7629f826f811c
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/_tabsyn_train.py
@@ -0,0 +1,63 @@
+import os, sys, subprocess
+
+work_dir = "/work/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916"
+dataname = "tabsyn_n16"
+tabsyn_root = "/workspace/tabsyn"
+
+assert os.path.exists(tabsyn_root), f"TabSyn source not mounted: {tabsyn_root}"
+
+old = os.environ.get("PYTHONPATH", "")
+os.environ["PYTHONPATH"] = tabsyn_root + (os.pathsep + old if old else "")
+sys.path.insert(0, tabsyn_root)
+
+os.chdir(tabsyn_root)
+
+# Symlink data dir into TabSyn data/
+data_link = os.path.join(tabsyn_root, "data", dataname)
+data_src = os.path.join(work_dir, "data", dataname)
+os.makedirs(os.path.join(tabsyn_root, "data"), exist_ok=True)
+if os.path.exists(data_link):
+    os.remove(data_link)
+os.symlink(data_src, data_link)
+
+env = os.environ.copy()
+env.setdefault("TABSYN_RESUME", "1")
+env.setdefault("TABSYN_VAE_BATCH_SIZE", "1024")
+_te = 1000
+if _te is not None:
+    env["TABSYN_VAE_EPOCHS"] = str(_te)
+    env["TABSYN_DIFFUSION_MAX_EPOCHS"] = str(max(_te + 1, 2))
+
+# Data preprocessing is done on the host side (_prepare_data_dir)
+# which creates .npy files, train/test CSVs, and info.json
+
+# Step 1: Train VAE (produces latent embeddings)
+print(f"[TabSyn] Step 1/2: Training VAE in {tabsyn_root}, dataname={dataname}")
+ret = subprocess.run(
+    [sys.executable, "main.py",
+     "--dataname", dataname,
+     "--mode", "train",
+     "--method", "vae",
+     "--gpu", "0"],
+    cwd=tabsyn_root,
+    env=env
+)
+if ret.returncode != 0:
+    print("[TabSyn] VAE training failed")
+    sys.exit(ret.returncode)
+
+# Step 2: Train diffusion model on latent space
+print(f"[TabSyn] Step 2/2: Training diffusion model")
+ret = subprocess.run(
+    [sys.executable, "main.py",
+     "--dataname", dataname,
+     "--mode", "train",
+     "--method", "tabsyn",
+     "--gpu", "0"],
+    cwd=tabsyn_root,
+    env=env
+)
+if ret.returncode != 0:
+    print("[TabSyn] Diffusion training failed")
+    sys.exit(ret.returncode)
+print("[TabSyn] Training complete (VAE + Diffusion)")
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_test.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..010b9b9a432ea415d8a994e8bfd064d4a83d555c
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad912bdeee70b2f0dcc962ecfee2edb74ccdc3743624cf91fe758fdf219a1068
+size 128
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_train.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..010b9b9a432ea415d8a994e8bfd064d4a83d555c
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_cat_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad912bdeee70b2f0dcc962ecfee2edb74ccdc3743624cf91fe758fdf219a1068
+size 128
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_test.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_train.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..9b8453d97777141d522e4ef992d402cf6540dda4
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/X_num_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b985d943ff2eb3776c1da61b5cfd77fb18b76ad3f6e2b916f8d1b9f9ec8635d5
+size 27341528
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/info.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/info.json
new file mode 100644
index 0000000000000000000000000000000000000000..c5146f578af2e0474b21e3f43edc5e89e798477b
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/info.json
@@ -0,0 +1,309 @@
+{
+  "name": "tabsyn_n16",
+  "task_type": "multiclass",
+  "n_num_features": 30,
+  "n_cat_features": 0,
+  "train_size": 227845,
+  "num_col_idx": [
+    0,
+    1,
+    2,
+    3,
+    4,
+    5,
+    6,
+    7,
+    8,
+    9,
+    10,
+    11,
+    12,
+    13,
+    14,
+    15,
+    16,
+    17,
+    18,
+    19,
+    20,
+    21,
+    22,
+    23,
+    24,
+    25,
+    26,
+    27,
+    28,
+    29
+  ],
+  "cat_col_idx": [],
+  "target_col_idx": [
+    30
+  ],
+  "column_names": [
+    "Time",
+    "V1",
+    "V2",
+    "V3",
+    "V4",
+    "V5",
+    "V6",
+    "V7",
+    "V8",
+    "V9",
+    "V10",
+    "V11",
+    "V12",
+    "V13",
+    "V14",
+    "V15",
+    "V16",
+    "V17",
+    "V18",
+    "V19",
+    "V20",
+    "V21",
+    "V22",
+    "V23",
+    "V24",
+    "V25",
+    "V26",
+    "V27",
+    "V28",
+    "Amount",
+    "Class"
+  ],
+  "train_num": 227845,
+  "test_num": 227845,
+  "header": 0,
+  "file_type": "csv",
+  "data_path": "data/tabsyn_n16/train.csv",
+  "test_path": null,
+  "idx_mapping": {
+    "0": 0,
+    "1": 1,
+    "2": 2,
+    "3": 3,
+    "4": 4,
+    "5": 5,
+    "6": 6,
+    "7": 7,
+    "8": 8,
+    "9": 9,
+    "10": 10,
+    "11": 11,
+    "12": 12,
+    "13": 13,
+    "14": 14,
+    "15": 15,
+    "16": 16,
+    "17": 17,
+    "18": 18,
+    "19": 19,
+    "20": 20,
+    "21": 21,
+    "22": 22,
+    "23": 23,
+    "24": 24,
+    "25": 25,
+    "26": 26,
+    "27": 27,
+    "28": 28,
+    "29": 29,
+    "30": 30
+  },
+  "inverse_idx_mapping": {
+    "0": 0,
+    "1": 1,
+    "2": 2,
+    "3": 3,
+    "4": 4,
+    "5": 5,
+    "6": 6,
+    "7": 7,
+    "8": 8,
+    "9": 9,
+    "10": 10,
+    "11": 11,
+    "12": 12,
+    "13": 13,
+    "14": 14,
+    "15": 15,
+    "16": 16,
+    "17": 17,
+    "18": 18,
+    "19": 19,
+    "20": 20,
+    "21": 21,
+    "22": 22,
+    "23": 23,
+    "24": 24,
+    "25": 25,
+    "26": 26,
+    "27": 27,
+    "28": 28,
+    "29": 29,
+    "30": 30
+  },
+  "idx_name_mapping": {
+    "0": "Time",
+    "1": "V1",
+    "2": "V2",
+    "3": "V3",
+    "4": "V4",
+    "5": "V5",
+    "6": "V6",
+    "7": "V7",
+    "8": "V8",
+    "9": "V9",
+    "10": "V10",
+    "11": "V11",
+    "12": "V12",
+    "13": "V13",
+    "14": "V14",
+    "15": "V15",
+    "16": "V16",
+    "17": "V17",
+    "18": "V18",
+    "19": "V19",
+    "20": "V20",
+    "21": "V21",
+    "22": "V22",
+    "23": "V23",
+    "24": "V24",
+    "25": "V25",
+    "26": "V26",
+    "27": "V27",
+    "28": "V28",
+    "29": "Amount",
+    "30": "Class"
+  },
+  "n_classes": 2,
+  "metadata": {
+    "columns": {
+      "0": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "1": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "2": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "3": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "4": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "5": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "6": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "7": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "8": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "9": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "10": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "11": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "12": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "13": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "14": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "15": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "16": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "17": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "18": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "19": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "20": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "21": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "22": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "23": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "24": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "25": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "26": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "27": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "28": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "29": {
+        "sdtype": "numerical",
+        "computer_representation": "Float"
+      },
+      "30": {
+        "sdtype": "categorical"
+      }
+    }
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/test.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/train.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_test.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_test.npy
new file mode 100644
index 0000000000000000000000000000000000000000..00c862056efab08c66d26387390bc15c4a3f8e60
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_test.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:881f946dbf19ef8fb07061b76f6eb6969e9f3975e49dd6277ff18a39133fa251
+size 1822888
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_train.npy b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_train.npy
new file mode 100644
index 0000000000000000000000000000000000000000..00c862056efab08c66d26387390bc15c4a3f8e60
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/data/tabsyn_n16/y_train.npy
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:881f946dbf19ef8fb07061b76f6eb6969e9f3975e49dd6277ff18a39133fa251
+size 1822888
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/gen_20260427_002515.log b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/gen_20260427_002515.log
new file mode 100644
index 0000000000000000000000000000000000000000..6b98a293fe3a0d82ec22234df29ec45b4fcb9818
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/gen_20260427_002515.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4227639b2ad9c459342a196a8e47cae5bf6baf3fb87eb90f119988b1859ba0d5
+size 679
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/input_snapshot.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..3381d4200b231ed0d19e0c2054beb58e085f3098
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/input_snapshot.json
@@ -0,0 +1,36 @@
+{
+  "dataset_id": "n16",
+  "model": "tabsyn",
+  "inputs": {
+    "train_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+      "exists": true,
+      "size": 120434969,
+      "sha256": "d33c0f51e8e82d110fc3787d448c55eb646bb006d4142775762e29d38619dab7"
+    },
+    "val_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+      "exists": true,
+      "size": 15054728,
+      "sha256": "ba8c9f03e07e66776ad749730d985bb6491b848595a0727e851448fc8c0a28ce"
+    },
+    "test_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv",
+      "exists": true,
+      "size": 15054431,
+      "sha256": "23f5d0e05e2e308538143a51e1ceb6f019680d1e92ef6c37626cc7774f22b45b"
+    },
+    "profile_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_profile.json",
+      "exists": true,
+      "size": 13450,
+      "sha256": "dc84ba22eb87a13dee40814eb0af363bbe651e48726c8f07b641f4e30f33a235"
+    },
+    "contract_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_contract_v1.json",
+      "exists": true,
+      "size": 16859,
+      "sha256": "3e28017c8292679d4e8b4eeca7a3c31c5496db891c09e8b15d30f57e2884b12e"
+    }
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..d391c1ee1422eb5ee4ce05e4391604265338d18a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,655 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "columns": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/public_gate_report.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..77d1958fecefe23b3791fd07b5a6eaa8296b1d05
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/public_gate_report.json
@@ -0,0 +1,37 @@
+{
+  "dataset_id": "n16",
+  "status": "pass",
+  "checks": [
+    {
+      "check_id": "PG001_csv_parse_ok",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG002_split_header_consistent",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG003_profile_header_match",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG004_missing_token_normalized",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG005_semantic_type_validated",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG006_target_defined_and_valid",
+      "status": "pass"
+    }
+  ],
+  "target_column": "Class",
+  "task_type": "classification",
+  "input_splits": {
+    "train": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+    "val": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+    "test": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..9b5123daf5058127875d645f7ff2df93cee4b38f
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/staged_input_manifest.json
@@ -0,0 +1,660 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/public_gate_report.json",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/runtime_result.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..db05d00f3639c6e74cf3b1f9656957c88be63af6
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/runtime_result.json
@@ -0,0 +1,15 @@
+{
+  "dataset_id": "n16",
+  "model": "tabsyn",
+  "run_id": "tabsyn-n16-20260426_220916",
+  "public_gate_status": "pass",
+  "adapter_ready_status": "pass",
+  "train_status": "success",
+  "generate_status": "success",
+  "reason_code": null,
+  "reason_detail": null,
+  "artifacts": {
+    "synthetic_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv",
+    "model_path": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/staged_features.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..4fa94dfbf395e3c2359e2e16c0488f3f0b698f00
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/staged_features.json
@@ -0,0 +1,157 @@
+[
+  {
+    "feature_name": "Time",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V1",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V2",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V3",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V4",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V5",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V6",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V7",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V8",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V9",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V10",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V11",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V12",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V13",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V14",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V15",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V16",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V17",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V18",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V19",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V20",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V21",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V22",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V23",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V24",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V25",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V26",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V27",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V28",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Amount",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Class",
+    "data_type": "binary",
+    "is_target": true
+  }
+]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_report.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..b273895f312203c603adb931b06393d36837b046
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_report.json
@@ -0,0 +1,7 @@
+{
+  "adapter_ready_status": "pass",
+  "adapter_fail_reason_code": null,
+  "adapter_fail_detail": null,
+  "adapter_transforms_applied": [],
+  "model_input_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/model_input_manifest.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_transforms_applied.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..0637a088a01e8ddab3bf3fa98dbe804cbde1a0dc
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/adapter_transforms_applied.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/model_input_manifest.json b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..6ba1dcbba8336ad2e9dfaefdc4073518c4b33989
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/staged/tabsyn/model_input_manifest.json
@@ -0,0 +1,662 @@
+{
+  "dataset_id": "n16",
+  "model": "tabsyn",
+  "target_column": "Class",
+  "task_type": "classification",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ],
+  "public_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/staged_input_manifest.json",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tabsyn/tabsyn-n16-20260426_220916/public_gate/public_gate_report.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/real.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/real.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/real.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/test.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/synthetic/tabsyn_n16/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv
new file mode 100644
index 0000000000000000000000000000000000000000..bedd1cb35d19aa20792947ec7bee7036b84b3768
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/tabsyn-n16-227845-20260427_002515.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c4aafe1526b17b776efd708c27ea61ad82ec79e2ca17af6dacf020700112ba4
+size 75483189
diff --git a/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/train_20260426_221025.log b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/train_20260426_221025.log
new file mode 100644
index 0000000000000000000000000000000000000000..212746726db9baf6f914c4e29d9ddfdfbf41bfdd
--- /dev/null
+++ b/SynthData0523/main/n16/tabsyn/tabsyn-n16-20260426_220916/train_20260426_221025.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27a0e6a472152db0f33fbc2388e3682c8aa7aa74254a7c95da3c0940940d1a5a
+size 9203438
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_generate.py b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_generate.py
new file mode 100644
index 0000000000000000000000000000000000000000..399063a5190ff548aea28ec8a078013c09cfbb34
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_generate.py
@@ -0,0 +1,5 @@
+from ctgan.synthesizers.tvae import TVAE
+model = TVAE.load("/work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt")
+samples = model.sample(227845)
+samples.to_csv("/work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv", index=False)
+print(f"[TVAE] Generated 227845 rows -> /work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv")
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_train.py b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_train.py
new file mode 100644
index 0000000000000000000000000000000000000000..923572adbf702c9b2ebcaa2d768042e171ae4aae
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/_tvae_train.py
@@ -0,0 +1,16 @@
+import json, sys
+import pandas as pd
+from ctgan.data import read_csv
+from ctgan.synthesizers.tvae import TVAE
+
+csv_path = "/work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv"
+meta_path = "/work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/tvae_metadata.json"
+save_path = "/work/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt"
+epochs = 300
+
+data, discrete_columns = read_csv(csv_path, meta_path, header=True, discrete=None)
+print(f"[TVAE] Training on {len(data)} rows, {len(data.columns)} cols, epochs={epochs}")
+model = TVAE(epochs=epochs, batch_size=500)
+model.fit(data, discrete_columns)
+model.save(save_path)
+print(f"[TVAE] Model saved -> {save_path}")
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260328_164845.log b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260328_164845.log
new file mode 100644
index 0000000000000000000000000000000000000000..50259c2ebd57edf2a72e74c303d9f0f3c8bb5f77
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260328_164845.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebfe09840f93dc6d8ae946bb8d10dfe76b5f2b171df7b1b0b1dfaca71127d89b
+size 129
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260330_070842.log b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260330_070842.log
new file mode 100644
index 0000000000000000000000000000000000000000..65ff02249348c78c0863a19525102567ee822671
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/gen_20260330_070842.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15ff30a2b570c739a68837a003c5fb1b2dd159085376b1f96a36b700e41eddf7
+size 133
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/input_snapshot.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/input_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..44db04427454d037f8eeb3196ebad663b92e814a
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/input_snapshot.json
@@ -0,0 +1,36 @@
+{
+  "dataset_id": "n16",
+  "model": "tvae",
+  "inputs": {
+    "train_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+      "exists": true,
+      "size": 120434969,
+      "sha256": "d33c0f51e8e82d110fc3787d448c55eb646bb006d4142775762e29d38619dab7"
+    },
+    "val_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+      "exists": true,
+      "size": 15054728,
+      "sha256": "ba8c9f03e07e66776ad749730d985bb6491b848595a0727e851448fc8c0a28ce"
+    },
+    "test_csv": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv",
+      "exists": true,
+      "size": 15054431,
+      "sha256": "23f5d0e05e2e308538143a51e1ceb6f019680d1e92ef6c37626cc7774f22b45b"
+    },
+    "profile_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_profile.json",
+      "exists": true,
+      "size": 13450,
+      "sha256": "dc84ba22eb87a13dee40814eb0af363bbe651e48726c8f07b641f4e30f33a235"
+    },
+    "contract_json": {
+      "path": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/artifacts/data_core/tabular/n16/n16-dataset_contract_v1.json",
+      "exists": true,
+      "size": 16859,
+      "sha256": "3e28017c8292679d4e8b4eeca7a3c31c5496db891c09e8b15d30f57e2884b12e"
+    }
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/train_20260328_053849.log b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/train_20260328_053849.log
new file mode 100644
index 0000000000000000000000000000000000000000..13e47a8140927bd8d4947f4dff86719055d42512
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/train_20260328_053849.log
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5195669c34fbd466e4710ee40b047fbf5afa2bd553627de9adbe0507b6e88df8
+size 174
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c2c6581500a60e99bde6cb5c936d7771f531ddad
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/models_300epochs/tvae_300epochs.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc4db3762c3e8d22362458849a4f3ad1129a48583333b962b473eb088f25c301
+size 4989356
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/normalized_schema_snapshot.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/normalized_schema_snapshot.json
new file mode 100644
index 0000000000000000000000000000000000000000..d391c1ee1422eb5ee4ce05e4391604265338d18a
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/normalized_schema_snapshot.json
@@ -0,0 +1,655 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "columns": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/public_gate_report.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/public_gate_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..77d1958fecefe23b3791fd07b5a6eaa8296b1d05
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/public_gate_report.json
@@ -0,0 +1,37 @@
+{
+  "dataset_id": "n16",
+  "status": "pass",
+  "checks": [
+    {
+      "check_id": "PG001_csv_parse_ok",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG002_split_header_consistent",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG003_profile_header_match",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG004_missing_token_normalized",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG005_semantic_type_validated",
+      "status": "pass"
+    },
+    {
+      "check_id": "PG006_target_defined_and_valid",
+      "status": "pass"
+    }
+  ],
+  "target_column": "Class",
+  "task_type": "classification",
+  "input_splits": {
+    "train": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-train.csv",
+    "val": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-val.csv",
+    "test": "/data/jialinzhang/SynthesizePipeline-server/DatasetNew/n16/n16-test.csv"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/staged_input_manifest.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/staged_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..e1019c909e83626127b9613ba276182ac19c496d
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/public_gate/staged_input_manifest.json
@@ -0,0 +1,660 @@
+{
+  "dataset_id": "n16",
+  "target_column": "Class",
+  "task_type": "classification",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/public_gate/public_gate_report.json",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ]
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/runtime_result.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/runtime_result.json
new file mode 100644
index 0000000000000000000000000000000000000000..a30956d333504d32a5f91d952c27801f143dd546
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/runtime_result.json
@@ -0,0 +1,14 @@
+{
+  "dataset_id": "n16",
+  "model": "tvae",
+  "run_id": "tvae-n16-20260328_053742",
+  "public_gate_status": "pass",
+  "adapter_ready_status": "pass",
+  "train_status": "skipped",
+  "generate_status": "success",
+  "reason_code": null,
+  "reason_detail": null,
+  "artifacts": {
+    "synthetic_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv"
+  }
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/staged_features.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/staged_features.json
new file mode 100644
index 0000000000000000000000000000000000000000..4fa94dfbf395e3c2359e2e16c0488f3f0b698f00
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/staged_features.json
@@ -0,0 +1,157 @@
+[
+  {
+    "feature_name": "Time",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V1",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V2",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V3",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V4",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V5",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V6",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V7",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V8",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V9",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V10",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V11",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V12",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V13",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V14",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V15",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V16",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V17",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V18",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V19",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V20",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V21",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V22",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V23",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V24",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V25",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V26",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V27",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "V28",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Amount",
+    "data_type": "continuous",
+    "is_target": false
+  },
+  {
+    "feature_name": "Class",
+    "data_type": "binary",
+    "is_target": true
+  }
+]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv
new file mode 100644
index 0000000000000000000000000000000000000000..cb4c5f2390e09a1ae9f2ca49caa3db30b90ee24e
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d69f73fdce5c2172678b0625c43058e7b6c8a991cee8c51c62b4e8f9200ddda9
+size 15025273
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv
new file mode 100644
index 0000000000000000000000000000000000000000..ab9a51ab9c902997393ab9225f0450777380d61a
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9803535990b2822a664c439816ab91c39aeedc38f99ade09a4347fe013876a3
+size 120653578
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d9e3a8f016b961c43651de14560f3c57ba65483f
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3abf3e2b22202b610a02bb43c40fc2751b5ff791d7c12fbc78ec217fdd5bebb2
+size 15025064
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_report.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..0260199cf626045b9a9f71cabbe192fa17a85b66
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_report.json
@@ -0,0 +1,7 @@
+{
+  "adapter_ready_status": "pass",
+  "adapter_fail_reason_code": null,
+  "adapter_fail_detail": null,
+  "adapter_transforms_applied": [],
+  "model_input_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/tvae/model_input_manifest.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_transforms_applied.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_transforms_applied.json
new file mode 100644
index 0000000000000000000000000000000000000000..0637a088a01e8ddab3bf3fa98dbe804cbde1a0dc
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/adapter_transforms_applied.json
@@ -0,0 +1 @@
+[]
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/model_input_manifest.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/model_input_manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..07005435bc3d5c6f24f70be6fb0b2a1f870f692f
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/staged/tvae/model_input_manifest.json
@@ -0,0 +1,662 @@
+{
+  "dataset_id": "n16",
+  "model": "tvae",
+  "target_column": "Class",
+  "task_type": "classification",
+  "column_schema": [
+    {
+      "name": "Time",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "71173",
+          "51740",
+          "43731",
+          "62140",
+          "54890"
+        ]
+      }
+    },
+    {
+      "name": "V1",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.61764107545646",
+          "0.114981070303704",
+          "-1.37134345385227",
+          "1.27608851046025",
+          "1.19879113285333"
+        ]
+      }
+    },
+    {
+      "name": "V2",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0712640179147479",
+          "-2.17505208918391",
+          "1.91649244679329",
+          "0.178601857699708",
+          "0.141561830103782"
+        ]
+      }
+    },
+    {
+      "name": "V3",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.08756433668229",
+          "0.5591762331232",
+          "-2.6152361559883",
+          "-0.435947524195021",
+          "0.128357399534487"
+        ]
+      }
+    },
+    {
+      "name": "V4",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.53757541136873",
+          "0.745309411772194",
+          "1.01424541859702",
+          "0.128762338296452",
+          "0.951394482083887"
+        ]
+      }
+    },
+    {
+      "name": "V5",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.05205150595571",
+          "-1.74600511198915",
+          "1.36889632754891",
+          "0.702271228090793",
+          "-0.214895831098192"
+        ]
+      }
+    },
+    {
+      "name": "V6",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.0978114651908514",
+          "0.141649291804806",
+          "-0.955056757624549",
+          "0.712898098074311",
+          "-0.487929084819731"
+        ]
+      }
+    },
+    {
+      "name": "V7",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.913248764299592",
+          "-0.0984133600931447",
+          "1.90583080967283",
+          "-0.124131417767229",
+          "0.00985233618123693"
+        ]
+      }
+    },
+    {
+      "name": "V8",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.43255733444336",
+          "0.125246368875306",
+          "-1.11209948135248",
+          "0.193930892662116",
+          "0.0285660683044692"
+        ]
+      }
+    },
+    {
+      "name": "V9",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-2.22583876169292",
+          "0.967740340241297",
+          "1.82396007167921",
+          "-0.104273933068917",
+          "0.00219911468230363"
+        ]
+      }
+    },
+    {
+      "name": "V10",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.34264080366605",
+          "-0.442492445854468",
+          "2.44412724472461",
+          "-0.157454526051497",
+          "0.228395224600428"
+        ]
+      }
+    },
+    {
+      "name": "V11",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.36193768297988",
+          "0.717035325285831",
+          "2.18728732193274",
+          "0.29733572607023",
+          "0.716944702795966"
+        ]
+      }
+    },
+    {
+      "name": "V12",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.169484054008359",
+          "0.716391895436279",
+          "-0.991146370190166",
+          "0.406569001410629",
+          "-0.138356592577646"
+        ]
+      }
+    },
+    {
+      "name": "V13",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.130002953664293",
+          "-1.04320949940957",
+          "-0.919797931613834",
+          "0.365479486184138",
+          "-1.72993685835762"
+        ]
+      }
+    },
+    {
+      "name": "V14",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.204792502243875",
+          "-0.0316449564066733",
+          "-4.83016981481996",
+          "-0.152449710027495",
+          "0.907707219744815"
+        ]
+      }
+    },
+    {
+      "name": "V15",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.620557629213784",
+          "-0.866006053568429",
+          "0.892980466209571",
+          "0.779477069791457",
+          "0.43201608388236"
+        ]
+      }
+    },
+    {
+      "name": "V16",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.339516249010229",
+          "0.234957102059586",
+          "0.830100219091388",
+          "0.890599961000214",
+          "0.424059743838334"
+        ]
+      }
+    },
+    {
+      "name": "V17",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.439372637095307",
+          "-0.0519580276301838",
+          "1.71374040959502",
+          "-0.602468223043572",
+          "-0.700215368047624"
+        ]
+      }
+    },
+    {
+      "name": "V18",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.0331053057534185",
+          "0.0965133524403355",
+          "2.07929445001711",
+          "0.320382698546599",
+          "0.663909190741365"
+        ]
+      }
+    },
+    {
+      "name": "V19",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-1.10630117422884",
+          "0.520211389792567",
+          "0.0719381386223041",
+          "0.382126293224404",
+          "0.0518179632547461"
+        ]
+      }
+    },
+    {
+      "name": "V20",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.656321350705548",
+          "0.966176191974396",
+          "0.836354777377543",
+          "-0.0325591024416928",
+          "-0.205023826129381"
+        ]
+      }
+    },
+    {
+      "name": "V21",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "1.43818296518693",
+          "0.135150072640343",
+          "-0.751995815524176",
+          "-0.317670513623694",
+          "0.0454833098773692"
+        ]
+      }
+    },
+    {
+      "name": "V22",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.288013810481786",
+          "-0.671916589231319",
+          "-0.241563753367306",
+          "-0.93070361305103",
+          "0.0314813784928002"
+        ]
+      }
+    },
+    {
+      "name": "V23",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.245959524495802",
+          "-0.397340082352452",
+          "-0.332694934708389",
+          "-0.0737934231105157",
+          "-0.143514791164631"
+        ]
+      }
+    },
+    {
+      "name": "V24",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.150969259002792",
+          "0.399051651198262",
+          "-0.857689814008883",
+          "-1.75198485180637",
+          "-0.0358779858512518"
+        ]
+      }
+    },
+    {
+      "name": "V25",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.310248896352418",
+          "-0.123950127949627",
+          "-0.296615260101376",
+          "0.338834704210441",
+          "0.63964805960102"
+        ]
+      }
+    },
+    {
+      "name": "V26",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "-0.392927869132217",
+          "0.839132432043748",
+          "-0.598705356084532",
+          "0.213187631136729",
+          "-0.320850766234866"
+        ]
+      }
+    },
+    {
+      "name": "V27",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.409241487139802",
+          "-0.144365502244215",
+          "-1.12856674164726",
+          "-0.0146174626202719",
+          "-0.00408114367990229"
+        ]
+      }
+    },
+    {
+      "name": "V28",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "0.100438703084057",
+          "0.0970715173096018",
+          "-1.17740216851208",
+          "0.00100028727777057",
+          "0.00538738847157751"
+        ]
+      }
+    },
+    {
+      "name": "Amount",
+      "role": "feature",
+      "semantic_type": "numeric",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "median",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 20000,
+        "unique_ratio": 0.087779,
+        "example_values": [
+          "15",
+          "550.2",
+          "89.99",
+          "9.99",
+          "17.23"
+        ]
+      }
+    },
+    {
+      "name": "Class",
+      "role": "target",
+      "semantic_type": "boolean",
+      "nullable": false,
+      "missing_tokens": [],
+      "parse_format": null,
+      "impute_strategy": "mode",
+      "profile_stats": {
+        "missing_rate": 0.0,
+        "unique_count": 2,
+        "unique_ratio": 9e-06,
+        "example_values": [
+          "0",
+          "1"
+        ]
+      }
+    }
+  ],
+  "public_manifest": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/public_gate/staged_input_manifest.json",
+  "train_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/train.csv",
+  "val_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/val.csv",
+  "test_csv": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/test.csv",
+  "features_json": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/staged/public/staged_features.json",
+  "public_gate_report": "/data/jialinzhang/SynthesizePipeline-server/output-SpecializedModels/n16/tvae/tvae-n16-20260328_053742/public_gate/public_gate_report.json"
+}
\ No newline at end of file
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-1000-20260328_164845.csv b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-1000-20260328_164845.csv
new file mode 100644
index 0000000000000000000000000000000000000000..3e80960da18d7730d63affbd5a9ce45148c88da7
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-1000-20260328_164845.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:598730669eed1467d938d86cc42e43116e5ebc86a53967765cdb822cf56f0959
+size 595905
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv
new file mode 100644
index 0000000000000000000000000000000000000000..d053751b72c1d1d356ccf362c56bbc7444f338cd
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae-n16-227845-20260330_070842.csv
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:085448e7bbf081278cd3bb53979d54c71cc946999004041d1c173d1df4753145
+size 135738979
diff --git a/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae_metadata.json b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae_metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..90c43ff9b425e4b0d81825520d4e1ed776e832d3
--- /dev/null
+++ b/SynthData0523/main/n16/tvae/tvae-n16-20260328_053742/tvae_metadata.json
@@ -0,0 +1,128 @@
+{
+  "columns": [
+    {
+      "name": "Time",
+      "type": "continuous"
+    },
+    {
+      "name": "V1",
+      "type": "continuous"
+    },
+    {
+      "name": "V2",
+      "type": "continuous"
+    },
+    {
+      "name": "V3",
+      "type": "continuous"
+    },
+    {
+      "name": "V4",
+      "type": "continuous"
+    },
+    {
+      "name": "V5",
+      "type": "continuous"
+    },
+    {
+      "name": "V6",
+      "type": "continuous"
+    },
+    {
+      "name": "V7",
+      "type": "continuous"
+    },
+    {
+      "name": "V8",
+      "type": "continuous"
+    },
+    {
+      "name": "V9",
+      "type": "continuous"
+    },
+    {
+      "name": "V10",
+      "type": "continuous"
+    },
+    {
+      "name": "V11",
+      "type": "continuous"
+    },
+    {
+      "name": "V12",
+      "type": "continuous"
+    },
+    {
+      "name": "V13",
+      "type": "continuous"
+    },
+    {
+      "name": "V14",
+      "type": "continuous"
+    },
+    {
+      "name": "V15",
+      "type": "continuous"
+    },
+    {
+      "name": "V16",
+      "type": "continuous"
+    },
+    {
+      "name": "V17",
+      "type": "continuous"
+    },
+    {
+      "name": "V18",
+      "type": "continuous"
+    },
+    {
+      "name": "V19",
+      "type": "continuous"
+    },
+    {
+      "name": "V20",
+      "type": "continuous"
+    },
+    {
+      "name": "V21",
+      "type": "continuous"
+    },
+    {
+      "name": "V22",
+      "type": "continuous"
+    },
+    {
+      "name": "V23",
+      "type": "continuous"
+    },
+    {
+      "name": "V24",
+      "type": "continuous"
+    },
+    {
+      "name": "V25",
+      "type": "continuous"
+    },
+    {
+      "name": "V26",
+      "type": "continuous"
+    },
+    {
+      "name": "V27",
+      "type": "continuous"
+    },
+    {
+      "name": "V28",
+      "type": "continuous"
+    },
+    {
+      "name": "Amount",
+      "type": "continuous"
+    },
+    {
+      "name": "Class",
+      "type": "categorical"
+    }
+  ]
+}
\ No newline at end of file