|
{ |
|
"_name_or_path": "MCG-NJU/videomae-large-finetuned-kinetics", |
|
"architectures": [ |
|
"VideoMAEForVideoClassification" |
|
], |
|
"attention_probs_dropout_prob": 0.0, |
|
"decoder_hidden_size": 512, |
|
"decoder_intermediate_size": 2048, |
|
"decoder_num_attention_heads": 8, |
|
"decoder_num_hidden_layers": 12, |
|
"hidden_act": "gelu", |
|
"hidden_dropout_prob": 0.0, |
|
"hidden_size": 1024, |
|
"id2label": { |
|
"0": "no_mopping", |
|
"1": "mopping" |
|
}, |
|
"image_size": 224, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 4096, |
|
"label2id": { |
|
"mopping": 1, |
|
"no_mopping": 0 |
|
}, |
|
"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": 24, |
|
"patch_size": 16, |
|
"problem_type": "regression", |
|
"qkv_bias": true, |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.29.2", |
|
"tubelet_size": 2, |
|
"use_mean_pooling": true |
|
} |
|
|