config.json

{
  "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": 768,
  "id2label": {
    "0": "CLS0",
    "1": "CLS1",
    "2": "CLS2",
    "3": "CLS3"
  },
  "image_size": 256,
  "initializer_range": 0.02,
  "label2id": {
    "CLS0": 0,
    "CLS1": 1,
    "CLS2": 2,
    "CLS3": 3
  },
  "layer_norm_eps": 1e-05,
  "mlp_ratio": 4.0,
  "model_type": "swinv2",
  "num_channels": 3,
  "num_heads": [
    4,
    8,
    16,
    32
  ],
  "num_layers": 4,
  "patch_size": 4,
  "path_norm": true,
  "pretrained_window_sizes": [
    0,
    0,
    0,
    0
  ],
  "qkv_bias": true,
  "torch_dtype": "float32",
  "transformers_version": "4.21.0.dev0",
  "use_absolute_embeddings": false,
  "window_size": 8
}