videomae-base-finetuned-kinetics-0326_final_duration4

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: 0.3843
• Accuracy: 0.8754

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
0.0049	0.01	687	0.6878	0.8185
0.0161	1.01	1374	0.5844	0.8088
0.6614	2.01	2061	0.3424	0.8857
0.5088	3.01	2748	0.9065	0.7650
0.2091	4.01	3435	0.9256	0.7626
0.009	5.01	4122	0.7178	0.8169
0.29	6.01	4809	0.7745	0.7212
0.7553	7.01	5496	0.7524	0.8088
0.0055	8.01	6183	0.7167	0.8152
0.5588	9.01	6870	1.0367	0.7682
0.6595	10.01	7557	1.1490	0.7820
0.6293	11.01	8244	1.0725	0.7455
0.5309	12.01	8931	0.9115	0.8096
0.0211	13.01	9618	1.1408	0.7609
0.0024	14.01	10305	1.0798	0.8006
0.2369	15.01	10992	1.2863	0.7763
0.0193	16.01	11679	1.5129	0.7601
0.0014	17.01	12366	1.1389	0.7804

Framework versions

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