config.json

{
  "_name_or_path": "facebook/detr-resnet-50",
  "activation_dropout": 0.0,
  "activation_function": "relu",
  "architectures": [
    "DetrForObjectDetection"
  ],
  "attention_dropout": 0.0,
  "auxiliary_loss": false,
  "backbone": "resnet50",
  "backbone_config": null,
  "backbone_kwargs": {
    "in_chans": 3,
    "out_indices": [
      1,
      2,
      3,
      4
    ]
  },
  "bbox_cost": 5,
  "bbox_loss_coefficient": 5,
  "class_cost": 1,
  "classifier_dropout": 0.0,
  "d_model": 256,
  "decoder_attention_heads": 8,
  "decoder_ffn_dim": 2048,
  "decoder_layerdrop": 0.0,
  "decoder_layers": 6,
  "dice_loss_coefficient": 1,
  "dilation": false,
  "dropout": 0.1,
  "encoder_attention_heads": 8,
  "encoder_ffn_dim": 2048,
  "encoder_layerdrop": 0.0,
  "encoder_layers": 6,
  "eos_coefficient": 0.1,
  "giou_cost": 2,
  "giou_loss_coefficient": 2,
  "id2label": {
    "0": "Person",
    "1": "Ear",
    "2": "Earmuffs",
    "3": "Face",
    "4": "Face-guard",
    "5": "Face-mask-medical",
    "6": "Foot",
    "7": "Tools",
    "8": "Glasses",
    "9": "Gloves",
    "10": "Helmet",
    "11": "Hands",
    "12": "Head",
    "13": "Medical-suit",
    "14": "Shoes",
    "15": "Safety-suit",
    "16": "Safety-vest"
  },
  "init_std": 0.02,
  "init_xavier_std": 1.0,
  "is_encoder_decoder": true,
  "label2id": {
    "Ear": 1,
    "Earmuffs": 2,
    "Face": 3,
    "Face-guard": 4,
    "Face-mask-medical": 5,
    "Foot": 6,
    "Glasses": 8,
    "Gloves": 9,
    "Hands": 11,
    "Head": 12,
    "Helmet": 10,
    "Medical-suit": 13,
    "Person": 0,
    "Safety-suit": 15,
    "Safety-vest": 16,
    "Shoes": 14,
    "Tools": 7
  },
  "mask_loss_coefficient": 1,
  "max_position_embeddings": 1024,
  "model_type": "detr",
  "num_channels": 3,
  "num_hidden_layers": 6,
  "num_queries": 100,
  "position_embedding_type": "sine",
  "scale_embedding": false,
  "torch_dtype": "float32",
  "transformers_version": "4.46.0.dev0",
  "use_pretrained_backbone": true,
  "use_timm_backbone": true
}