config.json

{
  "attention_probs_dropout_prob": 0.0,
  "audio_mask_ratio": 0.15,
  "audio_patch_size": [
    16,
    16
  ],
  "audio_size": [
    1024,
    128
  ],
  "decoder_hidden_size": 384,
  "decoder_intermediate_size": 1536,
  "decoder_num_attention_heads": 6,
  "decoder_num_hidden_layers": 4,
  "hidden_act": "gelu",
  "hidden_dropout_prob": 0.0,
  "hidden_size": 768,
  "id2label": {
    "0": "LABEL_0"
  },
  "image_size": 224,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "label2id": {
    "LABEL_0": 0
  },
  "layer_norm_eps": 1e-12,
  "model_type": "tvlt",
  "norm_pix_loss": true,
  "num_attention_heads": 12,
  "num_audio_channels": 1,
  "num_channels": 3,
  "num_frames": 8,
  "num_hidden_layers": 12,
  "patch_size": 16,
  "qkv_bias": true,
  "task_mae": true,
  "task_matching": true,
  "transformers_version": "4.25.0.dev0",
  "use_mean_pooling": true,
  "video_mask_ratio": 0.75
}