{
  "_name_or_path": "microsoft/swinv2-base-patch4-window8-256",
  "architectures": [
    "Swinv2ForImageClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "depths": [
    2,
    2,
    18,
    2
  ],
  "drop_path_rate": 0.1,
  "embed_dim": 128,
  "encoder_stride": 32,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 1024,
  "id2label": {
    "0": "a",
    "1": "ae",
    "10": "i",
    "11": "ja",
    "12": "ka",
    "13": "la",
    "14": "ma",
    "15": "na",
    "16": "nga",
    "17": "nya",
    "18": "o",
    "19": "pa",
    "2": "ba",
    "20": "qa",
    "21": "ra",
    "22": "sa",
    "23": "ta",
    "24": "u",
    "25": "va",
    "26": "wa",
    "27": "xa",
    "28": "ya",
    "29": "za",
    "3": "ca",
    "4": "da",
    "5": "e",
    "6": "eu",
    "7": "fa",
    "8": "ga",
    "9": "ha"
  },
  "image_size": 256,
  "initializer_range": 0.02,
  "label2id": {
    "a": "0",
    "ae": "1",
    "ba": "2",
    "ca": "3",
    "da": "4",
    "e": "5",
    "eu": "6",
    "fa": "7",
    "ga": "8",
    "ha": "9",
    "i": "10",
    "ja": "11",
    "ka": "12",
    "la": "13",
    "ma": "14",
    "na": "15",
    "nga": "16",
    "nya": "17",
    "o": "18",
    "pa": "19",
    "qa": "20",
    "ra": "21",
    "sa": "22",
    "ta": "23",
    "u": "24",
    "va": "25",
    "wa": "26",
    "xa": "27",
    "ya": "28",
    "za": "29"
  },
  "layer_norm_eps": 1e-05,
  "mlp_ratio": 4.0,
  "model_type": "swinv2",
  "num_channels": 3,
  "num_heads": [
    4,
    8,
    16,
    32
  ],
  "num_layers": 4,
  "out_features": [
    "stage4"
  ],
  "out_indices": [
    4
  ],
  "patch_size": 4,
  "path_norm": true,
  "pretrained_window_sizes": [
    0,
    0,
    0,
    0
  ],
  "problem_type": "single_label_classification",
  "qkv_bias": true,
  "stage_names": [
    "stem",
    "stage1",
    "stage2",
    "stage3",
    "stage4"
  ],
  "torch_dtype": "float32",
  "transformers_version": "4.40.2",
  "use_absolute_embeddings": false,
  "window_size": 8
}