config.json

{
  "_name_or_path": "apple/mobilevit-small",
  "architectures": [
    "MobileViTForImageClassification"
  ],
  "aspp_dropout_prob": 0.1,
  "aspp_out_channels": 256,
  "atrous_rates": [
    6,
    12,
    18
  ],
  "attention_probs_dropout_prob": 0.0,
  "classifier_dropout_prob": 0.1,
  "conv_kernel_size": 3,
  "expand_ratio": 4.0,
  "hidden_act": "silu",
  "hidden_dropout_prob": 0.1,
  "hidden_sizes": [
    144,
    192,
    240
  ],
  "id2label": {
    "0": "ain",
    "1": "aleff",
    "10": "haa",
    "11": "jeem",
    "12": "kaaf",
    "13": "khaa",
    "14": "laam",
    "15": "meem",
    "16": "none",
    "17": "nun",
    "18": "ra",
    "19": "saad",
    "2": "bb",
    "20": "seen",
    "21": "sheen",
    "22": "ta",
    "23": "taa",
    "24": "thaa",
    "25": "thal",
    "26": "waw",
    "27": "ya",
    "28": "zay",
    "3": "dal",
    "4": "dha",
    "5": "dhad",
    "6": "fa",
    "7": "gaaf",
    "8": "ghain",
    "9": "ha"
  },
  "image_size": 256,
  "initializer_range": 0.02,
  "label2id": {
    "ain": "0",
    "aleff": "1",
    "bb": "2",
    "dal": "3",
    "dha": "4",
    "dhad": "5",
    "fa": "6",
    "gaaf": "7",
    "ghain": "8",
    "ha": "9",
    "haa": "10",
    "jeem": "11",
    "kaaf": "12",
    "khaa": "13",
    "laam": "14",
    "meem": "15",
    "none": "16",
    "nun": "17",
    "ra": "18",
    "saad": "19",
    "seen": "20",
    "sheen": "21",
    "ta": "22",
    "taa": "23",
    "thaa": "24",
    "thal": "25",
    "waw": "26",
    "ya": "27",
    "zay": "28"
  },
  "layer_norm_eps": 1e-05,
  "mlp_ratio": 2.0,
  "model_type": "mobilevit",
  "neck_hidden_sizes": [
    16,
    32,
    64,
    96,
    128,
    160,
    640
  ],
  "num_attention_heads": 4,
  "num_channels": 3,
  "output_stride": 32,
  "patch_size": 2,
  "problem_type": "single_label_classification",
  "qkv_bias": true,
  "semantic_loss_ignore_index": 255,
  "torch_dtype": "float32",
  "transformers_version": "4.41.2"
}