{ "activation_dropout": 0.0, "activation_function": "silu", "architectures": [ "RfDetrModel" ], "attention_bias": true, "attention_dropout": 0.0, "auxiliary_loss": true, "backbone_config": { "apply_layernorm": true, "attention_probs_dropout_prob": 0.0, "drop_path_rate": 0.0, "hidden_act": "gelu", "hidden_dropout_prob": 0.0, "hidden_size": 32, "image_size": 256, "initializer_range": 0.02, "layer_norm_eps": 1e-06, "layerscale_value": 1.0, "mlp_ratio": 4, "model_type": "rf_detr_dinov2", "num_attention_heads": 2, "num_channels": 3, "num_hidden_layers": 4, "num_windows": 2, "out_features": [ "stage2", "stage3" ], "out_indices": [ 2, 3 ], "patch_size": 16, "qkv_bias": true, "reshape_hidden_states": true, "stage_names": [ "stem", "stage1", "stage2", "stage3", "stage4" ], "use_mask_token": true, "use_swiglu_ffn": false, "window_block_indexes": [ 0, 1 ] }, "bbox_cost": 5, "bbox_loss_coefficient": 5, "c2f_num_blocks": 3, "class_cost": 2, "class_loss_coefficient": 1, "d_model": 32, "decoder_activation_function": "relu", "decoder_cross_attention_heads": 4, "decoder_ffn_dim": 32, "decoder_layers": 2, "decoder_n_points": 4, "decoder_self_attention_heads": 2, "dice_loss_coefficient": 1, "disable_custom_kernels": true, "dropout": 0.0, "dtype": "float32", "eos_coefficient": 0.1, "focal_alpha": 0.25, "giou_cost": 2, "giou_loss_coefficient": 2, "group_detr": 2, "hidden_expansion": 0.5, "id2label": { "0": "LABEL_0", "1": "LABEL_1", "2": "LABEL_2", "3": "LABEL_3", "4": "LABEL_4" }, "init_std": 0.02, "intermediate_size": 1024, "label2id": { "LABEL_0": 0, "LABEL_1": 1, "LABEL_2": 2, "LABEL_3": 3, "LABEL_4": 4 }, "layer_norm_eps": 1e-05, "mask_class_loss_coefficient": 5.0, "mask_dice_loss_coefficient": 5.0, "mask_downsample_ratio": 4, "mask_loss_coefficient": 1, "mask_point_sample_ratio": 16, "model_type": "rf_detr", "num_feature_levels": 1, "num_queries": 10, "segmentation_head_activation_function": "gelu", "transformers_version": "5.16.0.dev0" }