Update tiny models for Florence2ForConditionalGeneration

config.json

@@ -0,0 +1,78 @@
+{
+  "architectures": [
+    "Florence2ForConditionalGeneration"
+  ],
+  "dtype": "float32",
+  "image_token_id": 4,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "model_type": "florence2",
+  "text_config": {
+    "activation_dropout": 0.1,
+    "activation_function": "gelu",
+    "attention_dropout": 0.1,
+    "bos_token_id": 0,
+    "classifier_dropout": 0.0,
+    "d_model": 16,
+    "decoder_attention_heads": 1,
+    "decoder_ffn_dim": 16,
+    "decoder_layerdrop": 0.0,
+    "decoder_layers": 1,
+    "decoder_start_token_id": 2,
+    "dropout": 0.1,
+    "encoder_attention_heads": 1,
+    "encoder_ffn_dim": 16,
+    "encoder_layerdrop": 0.0,
+    "encoder_layers": 1,
+    "eos_token_id": 2,
+    "forced_eos_token_id": 2,
+    "init_std": 0.02,
+    "is_decoder": false,
+    "max_position_embeddings": 200,
+    "model_type": "bart",
+    "pad_token_id": 1,
+    "scale_embedding": false,
+    "tie_word_embeddings": true,
+    "use_cache": true,
+    "vocab_size": 51290
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.16.0.dev0",
+  "vision_config": {
+    "activation_function": "gelu",
+    "depths": [
+      1
+    ],
+    "drop_path_rate": 0.1,
+    "embed_dim": [
+      16
+    ],
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "max_position_embeddings": 50,
+    "max_temporal_embeddings": 100,
+    "mlp_ratio": 4.0,
+    "model_type": "florence_vision",
+    "num_groups": [
+      1
+    ],
+    "num_heads": [
+      1
+    ],
+    "patch_padding": [
+      3
+    ],
+    "patch_prenorm": [
+      false
+    ],
+    "patch_size": [
+      7
+    ],
+    "patch_stride": [
+      4
+    ],
+    "projection_dim": 16,
+    "qkv_bias": true,
+    "window_size": 12
+  }
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "eos_token_id": 2,
+  "forced_eos_token_id": 2,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 1,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2cb702f81ff1ec27f341e340038af451e00f59a809b7b623b6ad907c4ab5b5e
+size 3386960

preprocessor_config.json

@@ -0,0 +1,32 @@
+{
+  "auto_map": {
+    "AutoProcessor": "processing_florence2.Florence2Processor"
+  },
+  "crop_size": {
+    "height": 768,
+    "width": 768
+  },
+  "do_center_crop": false,
+  "do_convert_rgb": null,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "CLIPImageProcessor",
+  "image_seq_length": 577,
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 768,
+    "width": 768
+  }
+}

tokenizer_config.json

@@ -0,0 +1,23 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "errors": "replace",
+  "image_token": "<image>",
+  "is_local": true,
+  "local_files_only": false,
+  "mask_token": "<mask>",
+  "model_max_length": 1024,
+  "model_specific_special_tokens": {
+    "image_token": "<image>"
+  },
+  "pad_token": "<pad>",
+  "processor_class": "Florence2Processor",
+  "sep_token": "</s>",
+  "tokenizer_class": "RobertaTokenizer",
+  "trim_offsets": true,
+  "unk_token": "<unk>"
+}