{
  "architectures": [
    "ViTForImageClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "dtype": "float32",
  "encoder_stride": 16,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "id2label": {
    "0": "Cell",
    "1": "Cell-Multi",
    "2": "Cracking",
    "3": "Diode",
    "4": "Diode-Multi",
    "5": "Hot-Spot",
    "6": "Hot-Spot-Multi",
    "7": "No-Anomaly",
    "8": "Offline-Module",
    "9": "Shadowing",
    "10": "Soiling",
    "11": "Vegetation"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "label2id": {
    "Cell": 0,
    "Cell-Multi": 1,
    "Cracking": 2,
    "Diode": 3,
    "Diode-Multi": 4,
    "Hot-Spot": 5,
    "Hot-Spot-Multi": 6,
    "No-Anomaly": 7,
    "Offline-Module": 8,
    "Shadowing": 9,
    "Soiling": 10,
    "Vegetation": 11
  },
  "layer_norm_eps": 1e-12,
  "model_type": "vit",
  "num_attention_heads": 12,
  "num_channels": 3,
  "num_hidden_layers": 12,
  "patch_size": 16,
  "pooler_act": "tanh",
  "pooler_output_size": 768,
  "qkv_bias": true,
  "transformers_version": "5.12.1"
}