{
  "_commit_hash": "480370a8aeeed9fc8d78837b4e94e5f936fe73f2",
  "_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,
  "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,
  "finetuning_task": "object-detection",
  "giou_cost": 2,
  "giou_loss_coefficient": 2,
  "id2label": {
    "0": "person",
    "1": "bicycle",
    "10": "fire hydrant",
    "11": "stop sign",
    "12": "parking meter",
    "13": "bench",
    "14": "bird",
    "15": "cat",
    "16": "dog",
    "17": "horse",
    "18": "sheep",
    "19": "cow",
    "2": "car",
    "20": "elephant",
    "21": "bear",
    "22": "zebra",
    "23": "giraffe",
    "24": "backpack",
    "25": "umbrella",
    "26": "handbag",
    "27": "tie",
    "28": "suitcase",
    "29": "frisbee",
    "3": "motorcycle",
    "30": "skis",
    "31": "snowboard",
    "32": "sports ball",
    "33": "kite",
    "34": "baseball bat",
    "35": "baseball glove",
    "36": "skateboard",
    "37": "surfboard",
    "38": "tennis racket",
    "39": "bottle",
    "4": "airplane",
    "40": "wine glass",
    "41": "cup",
    "42": "fork",
    "43": "knife",
    "44": "spoon",
    "45": "bowl",
    "46": "banana",
    "47": "apple",
    "48": "sandwich",
    "49": "orange",
    "5": "bus",
    "50": "broccoli",
    "51": "carrot",
    "52": "hot dog",
    "53": "pizza",
    "54": "donut",
    "55": "cake",
    "56": "chair",
    "57": "couch",
    "58": "potted plant",
    "59": "bed",
    "6": "train",
    "60": "dining table",
    "61": "toilet",
    "62": "tv",
    "63": "laptop",
    "64": "mouse",
    "65": "remote",
    "66": "keyboard",
    "67": "cell phone",
    "68": "microwave",
    "69": "oven",
    "7": "truck",
    "70": "toaster",
    "71": "sink",
    "72": "refrigerator",
    "73": "book",
    "74": "clock",
    "75": "vase",
    "76": "scissors",
    "77": "teddy bear",
    "78": "hair drier",
    "79": "toothbrush",
    "8": "boat",
    "9": "traffic light"
  },
  "init_std": 0.02,
  "init_xavier_std": 1.0,
  "is_encoder_decoder": true,
  "label2id": {
    "airplane": "4",
    "apple": "47",
    "backpack": "24",
    "banana": "46",
    "baseball bat": "34",
    "baseball glove": "35",
    "bear": "21",
    "bed": "59",
    "bench": "13",
    "bicycle": "1",
    "bird": "14",
    "boat": "8",
    "book": "73",
    "bottle": "39",
    "bowl": "45",
    "broccoli": "50",
    "bus": "5",
    "cake": "55",
    "car": "2",
    "carrot": "51",
    "cat": "15",
    "cell phone": "67",
    "chair": "56",
    "clock": "74",
    "couch": "57",
    "cow": "19",
    "cup": "41",
    "dining table": "60",
    "dog": "16",
    "donut": "54",
    "elephant": "20",
    "fire hydrant": "10",
    "fork": "42",
    "frisbee": "29",
    "giraffe": "23",
    "hair drier": "78",
    "handbag": "26",
    "horse": "17",
    "hot dog": "52",
    "keyboard": "66",
    "kite": "33",
    "knife": "43",
    "laptop": "63",
    "microwave": "68",
    "motorcycle": "3",
    "mouse": "64",
    "orange": "49",
    "oven": "69",
    "parking meter": "12",
    "person": "0",
    "pizza": "53",
    "potted plant": "58",
    "refrigerator": "72",
    "remote": "65",
    "sandwich": "48",
    "scissors": "76",
    "sheep": "18",
    "sink": "71",
    "skateboard": "36",
    "skis": "30",
    "snowboard": "31",
    "spoon": "44",
    "sports ball": "32",
    "stop sign": "11",
    "suitcase": "28",
    "surfboard": "37",
    "teddy bear": "77",
    "tennis racket": "38",
    "tie": "27",
    "toaster": "70",
    "toilet": "61",
    "toothbrush": "79",
    "traffic light": "9",
    "train": "6",
    "truck": "7",
    "tv": "62",
    "umbrella": "25",
    "vase": "75",
    "wine glass": "40",
    "zebra": "22"
  },
  "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": "4.7.0.dev0",
  "use_pretrained_backbone": true,
  "use_timm_backbone": true
}