| { |
| "architectures": [ |
| "CheXGroundModel" |
| ], |
| "attn_implementation": "flash_attention_2", |
| "bos_token_id": 1, |
| "eos_token_id": 2, |
| "grpa_cfg": { |
| "_attn_implementation_autoset": true, |
| "architectures": [ |
| "GRPA" |
| ], |
| "attn_kl_soft_alpha": 0.5, |
| "cls_weight": 0.0, |
| "ddetr_cfg": { |
| "_attn_implementation_autoset": true, |
| "architectures": [ |
| "CustomDDETRModel" |
| ], |
| "ddetr_cfg": { |
| "_attn_implementation_autoset": true, |
| "abnormality_num_classes": 14, |
| "abnormality_num_labels": 14, |
| "activation_dropout": 0.0, |
| "activation_function": "relu", |
| "anatomy_num_queries": 29, |
| "attention_dropout": 0.0, |
| "auxiliary_loss": true, |
| "backbone": "resnet50", |
| "backbone_config": null, |
| "backbone_kwargs": { |
| "in_chans": 3, |
| "out_indices": [ |
| 2, |
| 3, |
| 4 |
| ] |
| }, |
| "bbox_cost": 5, |
| "bbox_loss_coefficient": 5, |
| "class_cost": 1, |
| "d_model": 256, |
| "decoder_attention_heads": 8, |
| "decoder_ffn_dim": 1024, |
| "decoder_layers": 3, |
| "decoder_n_points": 8, |
| "detach_anatomy_boxes_for_abnormality": true, |
| "dice_loss_coefficient": 1, |
| "dilation": false, |
| "disable_custom_kernels": false, |
| "dropout": 0.1, |
| "encoder_attention_heads": 8, |
| "encoder_ffn_dim": 1024, |
| "encoder_layerdrop": 0.0, |
| "encoder_layers": 4, |
| "encoder_n_points": 4, |
| "eos_coefficient": 0.1, |
| "focal_alpha": 0.25, |
| "giou_cost": 2, |
| "giou_loss_coefficient": 5, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1", |
| "10": "LABEL_10", |
| "11": "LABEL_11", |
| "12": "LABEL_12", |
| "13": "LABEL_13", |
| "2": "LABEL_2", |
| "3": "LABEL_3", |
| "4": "LABEL_4", |
| "5": "LABEL_5", |
| "6": "LABEL_6", |
| "7": "LABEL_7", |
| "8": "LABEL_8", |
| "9": "LABEL_9" |
| }, |
| "init_std": 0.02, |
| "init_xavier_std": 1.0, |
| "is_encoder_decoder": true, |
| "kl_loss_weight": 0.1, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1, |
| "LABEL_10": 10, |
| "LABEL_11": 11, |
| "LABEL_12": 12, |
| "LABEL_13": 13, |
| "LABEL_2": 2, |
| "LABEL_3": 3, |
| "LABEL_4": 4, |
| "LABEL_5": 5, |
| "LABEL_6": 6, |
| "LABEL_7": 7, |
| "LABEL_8": 8, |
| "LABEL_9": 9 |
| }, |
| "mask_loss_coefficient": 1, |
| "max_position_embeddings": 1024, |
| "model_type": "deformable_detr", |
| "num_channels": 3, |
| "num_feature_levels": 3, |
| "num_queries": 29, |
| "position_embedding_type": "sine", |
| "roi_bce_weight": 0.01, |
| "teacher_bce_weight": 1.0, |
| "teacher_feature_source": "encoder", |
| "train_abnormality_roi": false, |
| "train_abnormality_teacher": true, |
| "train_bbox": true, |
| "transformers_version": "4.51.3", |
| "two_stage": false, |
| "two_stage_num_proposals": 29, |
| "use_pretrained_backbone": true, |
| "use_timm_backbone": true, |
| "with_box_refine": true |
| }, |
| "model_type": "ddetr", |
| "torch_dtype": "float32", |
| "transformers_version": "4.51.3", |
| "vis_encoder_cfg": { |
| "_attn_implementation_autoset": true, |
| "apply_layernorm": true, |
| "architectures": [ |
| "Dinov2Model" |
| ], |
| "attention_probs_dropout_prob": 0.0, |
| "drop_path_rate": 0.0, |
| "dtype": "float32", |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.0, |
| "hidden_size": 768, |
| "image_size": 518, |
| "initializer_range": 0.02, |
| "layer_norm_eps": 1e-06, |
| "layerscale_value": 1.0, |
| "mlp_ratio": 4, |
| "model_type": "dinov2", |
| "num_attention_heads": 12, |
| "num_channels": 3, |
| "num_hidden_layers": 12, |
| "out_features": [ |
| "stage12" |
| ], |
| "out_indices": [ |
| 12 |
| ], |
| "patch_size": 14, |
| "qkv_bias": true, |
| "reshape_hidden_states": true, |
| "stage_names": [ |
| "stem", |
| "stage1", |
| "stage2", |
| "stage3", |
| "stage4", |
| "stage5", |
| "stage6", |
| "stage7", |
| "stage8", |
| "stage9", |
| "stage10", |
| "stage11", |
| "stage12" |
| ], |
| "transformers_version": "4.51.3", |
| "use_mask_token": true, |
| "use_swiglu_ffn": false |
| }, |
| "vis_output_layer": -1, |
| "zs_weight_path": null |
| }, |
| "ddetr_checkpoint": null, |
| "global_contrastive_weight": 1.0, |
| "global_mlp_ratio": 4.0, |
| "global_num_heads": 8, |
| "jitter_expand_max": 0.3, |
| "jitter_scale": 0.05, |
| "jitter_translate": 0.0, |
| "local_aux_type": "attn_kl", |
| "local_aux_weight": 1.0, |
| "local_gloria_weight": 1.0, |
| "model_type": "grpa", |
| "num_classes": 14, |
| "roi_mlp_ratio": 4.0, |
| "roi_num_heads": 8, |
| "roi_output_size": 7, |
| "roi_sampling_ratio": 2, |
| "temp1": 4.0, |
| "temp2": 5.0, |
| "temp3": 15.0, |
| "temporal_mlp_ratio": null, |
| "temporal_num_heads": null, |
| "temporal_use_libra_prior_bias": true, |
| "text_encoder_cfg": { |
| "architectures": [ |
| "CXRBertModel" |
| ], |
| "attention_probs_dropout_prob": 0.25, |
| "auto_map": { |
| "AutoConfig": "configuration_cxrbert.CXRBertConfig", |
| "AutoModel": "modeling_cxrbert.CXRBertModel" |
| }, |
| "classifier_dropout": null, |
| "gradient_checkpointing": false, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.25, |
| "hidden_size": 768, |
| "initializer_range": 0.02, |
| "intermediate_size": 3072, |
| "layer_norm_eps": 1e-12, |
| "max_position_embeddings": 512, |
| "model_type": "cxr-bert", |
| "num_attention_heads": 12, |
| "num_hidden_layers": 12, |
| "pad_token_id": 0, |
| "position_embedding_type": "absolute", |
| "projection_size": 128, |
| "torch_dtype": "float32", |
| "transformers_version": "4.51.3", |
| "type_vocab_size": 2, |
| "use_cache": true, |
| "vocab_size": 30522 |
| }, |
| "text_encoder_name": "microsoft/BiomedVLP-BioViL-T", |
| "text_max_length": 96, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.51.3" |
| }, |
| "hidden_size": 4096, |
| "image_processor_name": "microsoft/rad-dino", |
| "libra_cfg": { |
| "architectures": [ |
| "LibraLlamaForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attn_implementation": "flash_attention_2", |
| "bos_token_id": 1, |
| "eos_token_id": 2, |
| "freeze_mm_mlp_adapter": true, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 4096, |
| "image_aspect_ratio": "pad", |
| "initializer_range": 0.02, |
| "intermediate_size": 11008, |
| "max_position_embeddings": 2048, |
| "mlp_bias": false, |
| "mm_hidden_size": 768, |
| "mm_projector_lr": 2e-05, |
| "mm_projector_type": "TAC", |
| "mm_use_im_patch_token": false, |
| "mm_use_im_start_end": false, |
| "mm_vision_select_feature": "patch", |
| "mm_vision_select_layer": "all", |
| "mm_vision_tower": "microsoft/rad-dino", |
| "model_type": "libra", |
| "num_attention_heads": 32, |
| "num_hidden_layers": 32, |
| "num_key_value_heads": 32, |
| "pretraining_tp": 1, |
| "rms_norm_eps": 1e-05, |
| "rope_scaling": null, |
| "rope_theta": 10000.0, |
| "tie_word_embeddings": false, |
| "tokenizer_model_max_length": 4096, |
| "tokenizer_padding_side": "right", |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.51.3", |
| "tune_mm_mlp_adapter": false, |
| "use_cache": false, |
| "use_mm_proj": true, |
| "vocab_size": 32017 |
| }, |
| "max_temporal_frames": 2, |
| "mm_hidden_size": 768, |
| "mm_projector_type": "TAC", |
| "mm_vision_tower": "microsoft/rad-dino", |
| "model_type": "chexground", |
| "num_new_token": 40, |
| "num_slots": 29, |
| "tie_word_embeddings": false, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.51.3", |
| "use_cache": false, |
| "vision_tower_override": "microsoft/rad-dino", |
| "vocab_size": 32057 |
| } |
|
|
