config.json

{
  "_name_or_path": "MCG-NJU/videomae-small-finetuned-kinetics",
  "architectures": [
    "VideoMAEForVideoClassification"
  ],
  "attention_probs_dropout_prob": 0.0,
  "decoder_hidden_size": 192,
  "decoder_intermediate_size": 768,
  "decoder_num_attention_heads": 3,
  "decoder_num_hidden_layers": 12,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 384,
  "id2label": {
    "0": "anh",
    "1": "hen gap lai",
    "2": "toi khoe",
    "3": "ong",
    "4": "ba",
    "5": "me",
    "6": "ban khoe k",
    "7": "bo",
    "8": "bac",
    "9": "cai gi",
    "10": "co",
    "11": "ban that tuyet",
    "12": "chu"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 1536,
  "label2id": {
    "anh": 0,
    "ba": 4,
    "bac": 8,
    "ban khoe k": 6,
    "ban that tuyet": 11,
    "bo": 7,
    "cai gi": 9,
    "chu": 12,
    "co": 10,
    "hen gap lai": 1,
    "me": 5,
    "ong": 3,
    "toi khoe": 2
  },
  "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": 12,
  "patch_size": 16,
  "problem_type": "single_label_classification",
  "qkv_bias": true,
  "torch_dtype": "float32",
  "transformers_version": "4.30.2",
  "tubelet_size": 2,
  "use_mean_pooling": true
}