{
  "_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": "Alhamdulillah",
    "1": "Good bye",
    "2": "Good evening",
    "3": "Good morning",
    "4": "How are you",
    "5": "I am fine",
    "6": "I am pleased to meet you",
    "7": "I am sorry",
    "8": "Not bad",
    "9": "Salam aleikum",
    "10": "Sorry",
    "11": "Thanks"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 1536,
  "label2id": {
    "Alhamdulillah": 0,
    "Good bye": 1,
    "Good evening": 2,
    "Good morning": 3,
    "How are you": 4,
    "I am fine": 5,
    "I am pleased to meet you": 6,
    "I am sorry": 7,
    "Not bad": 8,
    "Salam aleikum": 9,
    "Sorry": 10,
    "Thanks": 11
  },
  "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": "4.39.3",
  "tubelet_size": 2,
  "use_mean_pooling": true
}