Resume SynthData0523 main/c2 batch 21

.gitattributes

@@ -3775,3 +3775,50 @@ SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/run_config.json filte
 SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/runtime_result.json filter=lfs diff=lfs merge=lfs -text
 SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
 SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/tabpfgen-c2-1382-20260505_025721.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/tabpfgen_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/train_20260505_025721.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/gen_20260505_025826.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/run_config.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/runtime_result.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen-c2-1382-20260505_025826.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/train_20260505_025826.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/gen_20260505_025915.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/run_config.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/runtime_result.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/staged_features.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/test.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/train.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/val.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/adapter_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/adapter_transforms_applied.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/model_input_manifest.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen-c2-1382-20260505_025915.csv filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen_meta.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/train_20260505_025915.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/gen_20260505_030003.log filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/input_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/public_gate/normalized_schema_snapshot.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/public_gate/public_gate_report.json filter=lfs diff=lfs merge=lfs -text
+SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/public_gate/staged_input_manifest.json filter=lfs diff=lfs merge=lfs -text

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aed00c2c2b3f88a55a7ebff31b2e1b5e0e32fb0a7267e0b9d2779cd23e434dd
+size 41565

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/public/val.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26e90c1170a57a14c05832ac88027722b1f3848f9662c7c09ef7c93dcba4cc01
+size 5176

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/adapter_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f2da9e799c7e55650dbbae8e6c794a25ff9be8812fa443cb7d49cb02d11d729
+size 324

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/adapter_transforms_applied.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/staged/tabpfgen/model_input_manifest.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80cf0d64b5d17d7d47127b5f3ab58a1faa170b8db5fc252dcbf40a818455de3d
+size 4169

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/tabpfgen-c2-1382-20260505_025721.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64e0904eb3f6d3ed59f3b018d298ed93eabfc8e11b3dfd67f35dbea8cfdfc799
+size 41865

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/tabpfgen_meta.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf713c4152f1b71ff9f4bbb84f4659d70c6cb5e3c313e74c892ac35e46a84710
+size 444

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025721/train_20260505_025721.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb3fc7aeb3c9879e8f1df3f581842107c40765cefe2f4d100b40562c8fbc968c
+size 595

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/_tabpfgen_generate.py

@@ -0,0 +1,122 @@
+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/c2/tabpfgen/tabpfgen-c2-20260505_025826/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(1382)
+
+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"
+
+# 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=1000,
+    sgld_step_size=0.01,
+    sgld_noise_scale=0.01,
+    device=device,
+)
+
+print(f"[TabPFGen] Generating {target_n} rows via generate_classification (chunk_rows={chunk_rows}, device={device})")
+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/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen-c2-1382-20260505_025826.csv", index=False)
+print(f"[TabPFGen] Saved {len(syn_df)} rows -> /work/output-Benchmark-trainonly-v1/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen-c2-1382-20260505_025826.csv")

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/gen_20260505_025826.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd37470a04e5b77192013393ca5743e0dfaacf234072d5c2bdc4e8d5d8e400ec
+size 3226

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/input_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e348e38f04c3c1041d993b37fc998db9f8328874eca78bab60f77644edd5cb1a
+size 1344

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/normalized_schema_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a63ea2091eccf0145444f670415e7201dbdd052bc2ddabdc4241f6f16bdcf0a0
+size 3153

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/public_gate_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdc8c14ffce6055ca726f7949b2900c4ba619b670f8fc63c61328c24e4c46c66
+size 912

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/public_gate/staged_input_manifest.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5c682d794819c491053f9b673188e61444092bdb15712d2c7e753938a767ccb
+size 3969

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/run_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7fb730af63198afb935e198e426bc22c087c51333082ece0f505e6a61232eb6
+size 2087

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/runtime_result.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffaa4d78a653bd78190ea39b5fdc25245f9177e5c9bc9cf7bc27e3ca87293e83
+size 887

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/staged_features.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37511511f1d7cb0e004f9d7305c900644ebefe672c9a1a418c1f29b36786fabd
+size 659

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b48114a7d0bc5bd9a07920f903c8d4aba8bf98bf2a66a050da03588b0245ca73
+size 5273

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aed00c2c2b3f88a55a7ebff31b2e1b5e0e32fb0a7267e0b9d2779cd23e434dd
+size 41565

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/public/val.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26e90c1170a57a14c05832ac88027722b1f3848f9662c7c09ef7c93dcba4cc01
+size 5176

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/adapter_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07a415ebd159c95619e66abb7847eea48ec54fb35b4d273935764a4318e9e46c
+size 324

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/adapter_transforms_applied.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/staged/tabpfgen/model_input_manifest.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0e6177adb8ed1a42f1497d9d59bc34b24281d7f05abb6055bf077b49d23b86e
+size 4169

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen-c2-1382-20260505_025826.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b49af73dcfe0980b658726ad1d61f45776ec5edec34acf5c5a4babe5f81ab04
+size 41922

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/tabpfgen_meta.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5b10a119cb70b3375e3c29040e489f66f11ac27ebc8f44e8a1408af832bcd10
+size 444

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025826/train_20260505_025826.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a028185dfc74f2ea36397ef2db6c996b532a53ac12d317c0881382d013164fd3
+size 595

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/_tabpfgen_generate.py

@@ -0,0 +1,122 @@
+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/c2/tabpfgen/tabpfgen-c2-20260505_025915/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(1382)
+
+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"
+
+# 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=1000,
+    sgld_step_size=0.01,
+    sgld_noise_scale=0.01,
+    device=device,
+)
+
+print(f"[TabPFGen] Generating {target_n} rows via generate_classification (chunk_rows={chunk_rows}, device={device})")
+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/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen-c2-1382-20260505_025915.csv", index=False)
+print(f"[TabPFGen] Saved {len(syn_df)} rows -> /work/output-Benchmark-trainonly-v1/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen-c2-1382-20260505_025915.csv")

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/gen_20260505_025915.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f10d6172925223664b5cfe31493d3399422ab369b1702feb3f77793633057509
+size 3088

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/input_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e348e38f04c3c1041d993b37fc998db9f8328874eca78bab60f77644edd5cb1a
+size 1344

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/normalized_schema_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a63ea2091eccf0145444f670415e7201dbdd052bc2ddabdc4241f6f16bdcf0a0
+size 3153

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/public_gate_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdc8c14ffce6055ca726f7949b2900c4ba619b670f8fc63c61328c24e4c46c66
+size 912

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/public_gate/staged_input_manifest.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7f88af6efb77734ed84866cd4e9a6c34c720ccb632129ba2d74b2f1daa3ee91
+size 3969

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/run_config.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92f3fca043cf6f1e7146a5cfd9d0533b0f80f87ef23575132f80f256198fc542
+size 2087

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/runtime_result.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26955725007f292befbb8eaf1a1b1d5554a875343089f05b0485b023fe630c24
+size 887

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/staged_features.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37511511f1d7cb0e004f9d7305c900644ebefe672c9a1a418c1f29b36786fabd
+size 659

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/test.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b48114a7d0bc5bd9a07920f903c8d4aba8bf98bf2a66a050da03588b0245ca73
+size 5273

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/train.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4aed00c2c2b3f88a55a7ebff31b2e1b5e0e32fb0a7267e0b9d2779cd23e434dd
+size 41565

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/public/val.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26e90c1170a57a14c05832ac88027722b1f3848f9662c7c09ef7c93dcba4cc01
+size 5176

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/adapter_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ec6794fe19d39b5506585ce8e4ee1f4d4c22ed436c6871edf1a04f2d18f9ead
+size 324

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/adapter_transforms_applied.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f53cda18c2baa0c0354bb5f9a3ecbe5ed12ab4d8e11ba873c2f11161202b945
+size 2

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/staged/tabpfgen/model_input_manifest.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f8c96f2b6f6a7d757b4b57356199f4ba24e533aa6dfa7c6f9bfecec4465b723
+size 4169

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen-c2-1382-20260505_025915.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:504c5483efa154e46a03fc464f08465a925c9a3661cf315679c923e5e51beb4a
+size 41630

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/tabpfgen_meta.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e319dd44e41f80faf207c8443ccef38e79789b6d0498917e9689f4e0d21136e8
+size 444

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_025915/train_20260505_025915.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa2ea7b07a00948b6cef0740167b0b4d462754367e4bbd76bcd259eb1aaa8bd6
+size 595

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/_tabpfgen_generate.py

@@ -0,0 +1,122 @@
+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/c2/tabpfgen/tabpfgen-c2-20260505_030003/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(1382)
+
+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"
+
+# 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=1000,
+    sgld_step_size=0.01,
+    sgld_noise_scale=0.01,
+    device=device,
+)
+
+print(f"[TabPFGen] Generating {target_n} rows via generate_classification (chunk_rows={chunk_rows}, device={device})")
+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/c2/tabpfgen/tabpfgen-c2-20260505_030003/tabpfgen-c2-1382-20260505_030003.csv", index=False)
+print(f"[TabPFGen] Saved {len(syn_df)} rows -> /work/output-Benchmark-trainonly-v1/c2/tabpfgen/tabpfgen-c2-20260505_030003/tabpfgen-c2-1382-20260505_030003.csv")

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/gen_20260505_030003.log

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c89b6bed7c19b7b0a1370fc5740dccc0a448d2f0da2b81e17eccf9f13a9feb4c
+size 2300

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/input_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e348e38f04c3c1041d993b37fc998db9f8328874eca78bab60f77644edd5cb1a
+size 1344

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/public_gate/normalized_schema_snapshot.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a63ea2091eccf0145444f670415e7201dbdd052bc2ddabdc4241f6f16bdcf0a0
+size 3153

SynthData0523/main/c2/tabpfgen/tabpfgen-c2-20260505_030003/public_gate/public_gate_report.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdc8c14ffce6055ca726f7949b2900c4ba619b670f8fc63c61328c24e4c46c66
+size 912