config.json

{
  "architectures": [
    "SiglipForImageClassification"
  ],
  "id2label": {
    "0": "\u0905",
    "1": "\u0906",
    "2": "\u0907",
    "3": "\u0908",
    "4": "\u0909",
    "5": "\u090a",
    "6": "\u090f",
    "7": "\u0910",
    "8": "\u0913",
    "9": "\u0914",
    "10": "\u0915",
    "11": "\u0915\u094d\u0937",
    "12": "\u0916",
    "13": "\u0917",
    "14": "\u0918",
    "15": "\u091a",
    "16": "\u091b",
    "17": "\u091c",
    "18": "\u091c\u094d\u091e",
    "19": "\u091d",
    "20": "\u091f",
    "21": "\u0920",
    "22": "\u0921",
    "23": "\u0922",
    "24": "\u0923",
    "25": "\u0924",
    "26": "\u0925",
    "27": "\u0926",
    "28": "\u0927",
    "29": "\u0928",
    "30": "\u092a",
    "31": "\u092b",
    "32": "\u092c",
    "33": "\u092d",
    "34": "\u092e",
    "35": "\u092f",
    "36": "\u0930",
    "37": "\u0932",
    "38": "\u0933",
    "39": "\u0935",
    "40": "\u0936",
    "41": "\u0938",
    "42": "\u0939"
  },
  "initializer_factor": 1.0,
  "label2id": {
    "\u0905": 0,
    "\u0906": 1,
    "\u0907": 2,
    "\u0908": 3,
    "\u0909": 4,
    "\u090a": 5,
    "\u090f": 6,
    "\u0910": 7,
    "\u0913": 8,
    "\u0914": 9,
    "\u0915": 10,
    "\u0915\u094d\u0937": 11,
    "\u0916": 12,
    "\u0917": 13,
    "\u0918": 14,
    "\u091a": 15,
    "\u091b": 16,
    "\u091c": 17,
    "\u091c\u094d\u091e": 18,
    "\u091d": 19,
    "\u091f": 20,
    "\u0920": 21,
    "\u0921": 22,
    "\u0922": 23,
    "\u0923": 24,
    "\u0924": 25,
    "\u0925": 26,
    "\u0926": 27,
    "\u0927": 28,
    "\u0928": 29,
    "\u092a": 30,
    "\u092b": 31,
    "\u092c": 32,
    "\u092d": 33,
    "\u092e": 34,
    "\u092f": 35,
    "\u0930": 36,
    "\u0932": 37,
    "\u0933": 38,
    "\u0935": 39,
    "\u0936": 40,
    "\u0938": 41,
    "\u0939": 42
  },
  "model_type": "siglip",
  "problem_type": "single_label_classification",
  "text_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 768,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-06,
    "max_position_embeddings": 64,
    "model_type": "siglip_text_model",
    "num_attention_heads": 12,
    "num_hidden_layers": 12,
    "projection_size": 768,
    "torch_dtype": "float32",
    "vocab_size": 256000
  },
  "torch_dtype": "float32",
  "transformers_version": "4.50.0",
  "vision_config": {
    "attention_dropout": 0.0,
    "hidden_act": "gelu_pytorch_tanh",
    "hidden_size": 768,
    "image_size": 224,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-06,
    "model_type": "siglip_vision_model",
    "num_attention_heads": 12,
    "num_channels": 3,
    "num_hidden_layers": 12,
    "patch_size": 16,
    "torch_dtype": "float32"
  }
}