videomae-base-finetuned-slrbd002

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: 0.4824
• Accuracy: 0.875

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1800

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.6684	0.05	90	3.6696	0.025
3.4781	1.05	180	3.3494	0.1313
3.2885	2.05	270	3.1702	0.125
2.7847	3.05	360	2.7654	0.2562
2.4107	4.05	450	2.6756	0.2437
1.7783	5.05	540	1.9950	0.4813
1.3054	6.05	630	1.4908	0.6188
0.5868	7.05	720	1.4568	0.6188
0.4388	8.05	810	0.9511	0.7312
0.2187	9.05	900	1.1580	0.65
0.1665	10.05	990	0.6565	0.8313
0.0959	11.05	1080	0.5731	0.8562
0.0273	12.05	1170	0.6637	0.8063
0.02	13.05	1260	0.5048	0.875
0.0137	14.05	1350	0.4815	0.8688
0.0261	15.05	1440	0.5649	0.8438
0.013	16.05	1530	0.5419	0.8375
0.0123	17.05	1620	0.4864	0.8812
0.0527	18.05	1710	0.4725	0.8875
0.011	19.05	1800	0.4824	0.875

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0