videoMAE_base_wlasl_100_30ep_coR

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: 4.4320
• Accuracy: 0.0237

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: 5400
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
18.6471	0.0333	180	4.6379	0.0118
18.6042	1.0333	360	4.6276	0.0178
18.5574	2.0332	540	4.6139	0.0207
18.3992	3.0334	721	4.6085	0.0266
18.4284	4.0333	901	4.6058	0.0266
18.2402	5.0333	1081	4.6119	0.0266
18.1026	6.0332	1261	4.6158	0.0178
17.6285	7.0334	1442	4.5742	0.0266
17.0193	8.0333	1622	4.4320	0.0237

Framework versions

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