config.json

{
    "model_type": "tvp",
    "architectures": [
      "TvpForVideoGrounding"
    ],
    "distance_loss_weight": 1.0,
    "duration_loss_weight": 0.1,
    "visual_prompter_type": "framepad",
    "visual_prompter_apply": "replace",
    "max_img_size": 448,
    "visual_prompt_size": 96,
    "num_frames": 64,
    "attention_probs_dropout_prob": 0.1,
    "hidden_size": 768,
    "classifier": "mlp",
    "classifier_dropout": null,
    "cls_hidden_scale": 2,
    "hidden_act": "gelu",
    "hidden_dropout_prob": 0.1,
    "initializer_range": 0.02,
    "intermediate_size": 3072,
    "layer_norm_eps": 1e-12,
    "max_grid_col_position_embeddings": 100,
    "max_grid_row_position_embeddings": 100,
    "max_position_embeddings": 512,
    "num_attention_heads": 12,
    "num_hidden_layers": 12,
    "pad_token_id": 0,
    "position_embedding_type": "absolute",
    "transformers_version": "4.29.2",
    "type_vocab_size": 2,
    "use_cache": true,
    "vocab_size": 30522,
    "max_text_length": 100,
    "backbone_config": {
      "architectures": [
        "ResNetModel"
      ],
      "depths": [
        3,
        4,
        6,
        3
      ],
      "downsample_in_first_stage": false,
      "embedding_size": 64,
      "hidden_act": "relu",
      "hidden_sizes": [
        256,
        512,
        1024,
        2048
      ],
      "layer_type": "bottleneck",
      "model_type": "resnet",
      "num_channels": 3,
      "torch_dtype": "float32",
      "downsample_in_bottleneck": true
    }
}