VideoMAE_WLASL_2000__200_epoch_p20_longtail

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 4.6155
• Accuracy: 0.0207
• Top 1 Accuracy: 0.0207
• Top 5 Accuracy: 0.0592
• Top 10 Accuracy: 0.1124
• Macro Precision: 0.0005
• Macro Recall: 0.0117
• Macro F1: 0.0010
• Pearson Corr: nan
• Spearman Corr: nan

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 180
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Top 1 Accuracy	Top 5 Accuracy	Top 10 Accuracy	Macro Precision	Macro Recall	Macro F1	Pearson Corr	Spearman Corr
18.6831	1.0	180	4.6155	0.0207	0.0207	0.0592	0.1124	0.0005	0.0117	0.0010	nan	nan

Framework versions

• Transformers 4.46.1
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.1