videomae-base-finetuned-kinetics-0326_final_patience

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

• Loss: 0.6152
• Accuracy: 0.8422

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: 8
• eval_batch_size: 8
• seed: 42
• 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: 0.05
• training_steps: 68700

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0032	0.01	687	0.8430	0.8112
0.0067	1.01	1374	0.8257	0.7844
0.0193	2.01	2061	0.3668	0.8720
0.1767	3.01	2748	0.7760	0.7966
1.1229	4.01	3435	0.7641	0.8079
0.0123	5.01	4122	0.3614	0.8825
0.0299	6.01	4809	1.0045	0.7593
0.31	7.01	5496	0.8982	0.8290
0.0079	8.01	6183	0.6962	0.8525
0.2623	9.01	6870	1.7303	0.6807
0.4076	10.01	7557	0.9215	0.7998
0.1049	11.01	8244	1.0585	0.7966
0.0079	12.01	8931	1.0430	0.8039
0.0366	13.01	9618	0.6344	0.8590
0.0007	14.01	10305	0.8699	0.8379
0.2756	15.01	10992	0.9275	0.8217

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0