config.json

{
  "_name_or_path": "MCG-NJU/videomae-large",
  "architectures": [
    "VideoMAEForVideoClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "decoder_hidden_size": 512,
  "decoder_intermediate_size": 2048,
  "decoder_num_attention_heads": 8,
  "decoder_num_hidden_layers": 12,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 1024,
  "id2label": {
    "0": "climb",
    "1": "crawl",
    "2": "grasp",
    "3": "hiding",
    "4": "jump",
    "5": "pick_up",
    "6": "pull",
    "7": "push",
    "8": "put_down",
    "9": "roll",
    "10": "running",
    "11": "slide",
    "12": "walking",
    "13": "zibaroon"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 4096,
  "label2id": {
    "climb": 0,
    "crawl": 1,
    "grasp": 2,
    "hiding": 3,
    "jump": 4,
    "pick_up": 5,
    "pull": 6,
    "push": 7,
    "put_down": 8,
    "roll": 9,
    "running": 10,
    "slide": 11,
    "walking": 12,
    "zibaroon": 13
  },
  "layer_norm_eps": 1e-12,
  "model_type": "videomae",
  "norm_pix_loss": true,
  "num_attention_heads": 16,
  "num_channels": 3,
  "num_frames": 16,
  "num_hidden_layers": 24,
  "patch_size": 16,
  "problem_type": "single_label_classification",
  "qkv_bias": true,
  "torch_dtype": "float32",
  "transformers_version": "4.43.3",
  "tubelet_size": 2,
  "use_mean_pooling": false
}