{
  "architectures": [
    "EomtForUniversalSegmentation"
  ],
  "attention_dropout": 0.0,
  "class_weight": 2.0,
  "dice_weight": 5.0,
  "drop_path_rate": 0.0,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 1024,
  "id2label": {
    "0": "road",
    "1": "sidewalk",
    "2": "building",
    "3": "wall",
    "4": "fence",
    "5": "pole",
    "6": "traffic light",
    "7": "traffic sign",
    "8": "vegetation",
    "9": "terrain",
    "10": "sky",
    "11": "person",
    "12": "rider",
    "13": "car",
    "14": "truck",
    "15": "bus",
    "16": "train",
    "17": "motorcycle",
    "18": "bicycle"
  },
  "image_size": 1024,
  "importance_sample_ratio": 0.75,
  "initializer_range": 0.02,
  "label2id": {
    "bicycle": 18,
    "building": 2,
    "bus": 15,
    "car": 13,
    "fence": 4,
    "motorcycle": 17,
    "person": 11,
    "pole": 5,
    "rider": 12,
    "road": 0,
    "sidewalk": 1,
    "sky": 10,
    "terrain": 9,
    "traffic light": 6,
    "traffic sign": 7,
    "train": 16,
    "truck": 14,
    "vegetation": 8,
    "wall": 3
  },
  "layer_norm_eps": 1e-06,
  "layerscale_value": 1e-05,
  "mask_weight": 5.0,
  "mlp_ratio": 4,
  "model_type": "eomt",
  "no_object_weight": 0.1,
  "num_attention_heads": 16,
  "num_blocks": 4,
  "num_channels": 3,
  "num_hidden_layers": 24,
  "num_queries": 100,
  "num_register_tokens": 4,
  "num_upscale_blocks": 2,
  "oversample_ratio": 3.0,
  "patch_size": 16,
  "torch_dtype": "float32",
  "train_num_points": 12544,
  "transformers_version": "4.54.0.dev0",
  "use_swiglu_ffn": false
}