config.json

{
  "_name_or_path": "nvidia/mit-b0",
  "architectures": [
    "SegformerForSemanticSegmentation"
  ],
  "attention_probs_dropout_prob": 0.0,
  "classifier_dropout_prob": 0.1,
  "decoder_hidden_size": 256,
  "depths": [
    2,
    2,
    2,
    2
  ],
  "downsampling_rates": [
    1,
    4,
    8,
    16
  ],
  "drop_path_rate": 0.1,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_sizes": [
    32,
    64,
    160,
    256
  ],
  "id2label": {
    "0": "sky",
    "1": "pole/structural object",
    "2": "pole/structural object",
    "3": "road",
    "4": "static",
    "5": "static",
    "6": "building",
    "7": "sidewalk",
    "8": "road boundaries",
    "9": "crosswalks",
    "10": "pole/structural object",
    "11": "pedestrian",
    "12": "dynamic",
    "13": "pole/structural object",
    "14": "pole/structural object",
    "15": "traffic light",
    "16": "pole/structural object",
    "17": "pole/structural object",
    "18": "lane",
    "19": "lane",
    "20": "pole/structural object",
    "21": "lane",
    "22": "road marking",
    "23": "road marking",
    "24": "traffic light",
    "25": "static",
    "26": "static",
    "27": "static",
    "28": "static",
    "29": "static",
    "30": "static",
    "31": "static",
    "32": "static",
    "33": "static",
    "34": "static",
    "35": "static",
    "36": "road boundaries",
    "37": "crosswalks",
    "38": "static",
    "39": "static",
    "40": "traffic light",
    "41": "pole/structural object",
    "42": "pole/structural object",
    "43": "static",
    "44": "traffic light",
    "45": "traffic light",
    "46": "traffic light",
    "47": "static",
    "48": "static",
    "49": "static",
    "50": "static",
    "51": "pole/structural object",
    "52": "static",
    "53": "static",
    "54": "static",
    "55": "pole/structural object",
    "56": "static",
    "57": "pole/structural object",
    "58": "static",
    "59": "building",
    "60": "sidewalk",
    "61": "road boundaries",
    "62": "pole/structural object",
    "63": "pole/structural object",
    "64": "pedestrian",
    "65": "pedestrian",
    "66": "road marking",
    "67": "lane",
    "68": "crosswalks",
    "69": "vehicle",
    "70": "vehicle",
    "71": "vehicle",
    "72": "traffic sign"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "label2id": {
    "building": 1,
    "crosswalks": 2,
    "dynamic": 3,
    "lane": 4,
    "pedestrian": 5,
    "pole/structural object": 6,
    "road": 7,
    "road boundaries": 8,
    "road marking": 9,
    "sidewalk": 10,
    "sky": 11,
    "static": 12,
    "traffic light": 13,
    "traffic sign": 14,
    "unlabeled": 0,
    "vehicle": 15
  },
  "layer_norm_eps": 1e-06,
  "mlp_ratios": [
    4,
    4,
    4,
    4
  ],
  "model_type": "segformer",
  "num_attention_heads": [
    1,
    2,
    5,
    8
  ],
  "num_channels": 3,
  "num_encoder_blocks": 4,
  "patch_sizes": [
    7,
    3,
    3,
    3
  ],
  "reshape_last_stage": true,
  "semantic_loss_ignore_index": 255,
  "sr_ratios": [
    8,
    4,
    2,
    1
  ],
  "strides": [
    4,
    2,
    2,
    2
  ],
  "torch_dtype": "float32",
  "transformers_version": "4.42.4"
}