{
  "_name_or_path": "shauray/ViTPose",
  "architectures": [
    "ViTPoseForImageClassification"
  ],
  "img_size": [256, 192],
  "patch_size": 16,
  "embed_dim": 384,
  "depth": 12,
  "num_heads": 12,
  "ratio": 1,
  "use_checkpoint": false,
  "mlp_ratio": 4,
  "qkv_bias": true,
  "drop_path_rate": 0.1,
  "keypoint_in_channels": 382,
  "keypoint_num_deconv_layers": 2,
  "keypoint_num_deconv_filters": [256, 256],
  "keypoint_num_deconv_kernels": [4, 4],
  "dropout_p": 0.0,
  "num_output_channels": 17
}