{ "add_projection": false, "architectures": [ "DPTForDepthEstimation" ], "attention_probs_dropout_prob": 0.0, "auxiliary_loss_weight": 0.4, "backbone": null, "backbone_config": { "add_fpn": false, "attention_probs_dropout_prob": 0.0, "auxiliary_channels": 256, "auxiliary_concat_input": false, "auxiliary_loss_weight": 0.4, "auxiliary_num_convs": 1, "drop_path_rate": 0.1, "hidden_act": "gelu", "hidden_dropout_prob": 0.0, "hidden_size": 1024, "image_size": 384, "initializer_range": 0.02, "intermediate_size": 4096, "layer_norm_eps": 1e-12, "layer_scale_init_value": 0.1, "model_type": "beit", "num_attention_heads": 16, "num_channels": 3, "num_hidden_layers": 24, "out_features": [ "stage6", "stage12", "stage18", "stage24" ], "out_indices": [ 6, 12, 18, 24 ], "patch_size": 16, "pool_scales": [ 1, 2, 3, 6 ], "reshape_hidden_states": false, "semantic_loss_ignore_index": 255, "stage_names": [ "stem", "stage1", "stage2", "stage3", "stage4", "stage5", "stage6", "stage7", "stage8", "stage9", "stage10", "stage11", "stage12", "stage13", "stage14", "stage15", "stage16", "stage17", "stage18", "stage19", "stage20", "stage21", "stage22", "stage23", "stage24" ], "use_absolute_position_embeddings": false, "use_auxiliary_head": true, "use_mask_token": false, "use_mean_pooling": true, "use_relative_position_bias": true, "use_shared_relative_position_bias": false, "vocab_size": 8192 }, "backbone_featmap_shape": null, "backbone_kwargs": null, "backbone_out_indices": null, "fusion_hidden_size": 256, "head_in_index": -1, "hidden_act": "gelu", "hidden_dropout_prob": 0.0, "hidden_size": 768, "image_size": 384, "initializer_range": 0.02, "intermediate_size": 3072, "is_hybrid": false, "layer_norm_eps": 1e-12, "model_type": "dpt", "neck_hidden_sizes": [ 256, 512, 1024, 1024 ], "neck_ignore_stages": [], "num_attention_heads": 12, "num_channels": 3, "num_hidden_layers": 12, "output_attentions": false, "patch_size": 16, "pooler_act": "tanh", "pooler_output_size": 768, "qkv_bias": true, "readout_type": "project", "reassemble_factors": [ 4, 2, 1, 0.5 ], "semantic_classifier_dropout": 0.1, "semantic_loss_ignore_index": 255, "torch_dtype": "float32", "transformers_version": null, "use_autobackbone": true, "use_auxiliary_head": true, "use_batch_norm_in_fusion_residual": false, "use_bias_in_fusion_residual": null, "use_pretrained_backbone": false, "use_timm_backbone": false }