{ "_name_or_path": "MCG-NJU/videomae-base", "architectures": [ "VideoMAEForVideoClassification" ], "attention_probs_dropout_prob": 0.0, "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": "B0A", "1": "B0B", "2": "D0X", "3": "G01", "4": "G02", "5": "G03", "6": "G04", "7": "G05", "8": "G06", "9": "G07", "10": "G08", "11": "G09", "12": "G10", "13": "G11" }, "image_size": 224, "initializer_range": 0.02, "intermediate_size": 3072, "label2id": { "B0A": 0, "B0B": 1, "D0X": 2, "G01": 3, "G02": 4, "G03": 5, "G04": 6, "G05": 7, "G06": 8, "G07": 9, "G08": 10, "G09": 11, "G10": 12, "G11": 13 }, "layer_norm_eps": 1e-12, "model_type": "videomae", "norm_pix_loss": true, "num_attention_heads": 12, "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": "4.47.1", "tubelet_size": 2, "use_mean_pooling": false }