config.json

{
    "_name_or_path": "./",
    "model_type": "VideoMAEv2_Base",
    "architectures": [
      "VideoMAEv2_Base"
    ],
    "auto_map": {
        "AutoModel": "modeling_videomaev2.VideoMAEv2",
        "AutoConfig": "modeling_config.VideoMAEv2Config"
    },
    "model_config":{
      "img_size": 224,
      "patch_size": 16,
      "in_chans": 3,
      "num_classes": 0,
      "embed_dim": 1280,
      "depth": 32,
      "num_heads": 16,
      "mlp_ratio": 4,
      "qkv_bias": true,
      "qk_scale": null,
      "drop_rate": 0.0,
      "attn_drop_rate": 0.0,
      "drop_path_rate": 0.0,
      "norm_layer": "nn.LayerNorm",
      "layer_norm_eps": 1e-6,
      "init_values": 0.0,
      "use_learnable_pos_emb": false,
      "tubelet_size": 2,
      "use_mean_pooling": true,
      "with_cp": false,
      "num_frames": 16,  
      "cos_attn": false
    },
    "transformers_version": "4.38.0",
    "use_cache": true
  }