{
  "_name_or_path": "/datasets/work/hb-mlaifsp-mm/work/checkpoints/uniformer_base_tl_384",
  "architectures": [
    "UniFormerModel"
  ],
  "attention_probs_dropout_prob": 0.0,
  "attn_drop_rate": 0.0,
  "auto_map": {
    "AutoModel": "modelling_uniformer.UniFormerModel"
  },
  "conv_stem": false,
  "depth": [
    5,
    8,
    20,
    7
  ],
  "drop_path_rate": 0.3,
  "drop_rate": 0.0,
  "embed_dim": [
    64,
    128,
    320,
    512
  ],
  "encoder_stride": 16,
  "head_dim": 64,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "image_size": 384,
  "in_chans": 3,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "layer_norm_eps": 1e-06,
  "mlp_ratio": 4,
  "model_type": "vit",
  "num_attention_heads": 12,
  "num_channels": 3,
  "num_classes": 1000,
  "num_hidden_layers": 12,
  "patch_size": [
    4,
    2,
    2,
    2
  ],
  "projection_size": null,
  "qk_scale": null,
  "qkv_bias": true,
  "representation_size": null,
  "torch_dtype": "float32",
  "transformers_version": "4.40.2"
}