videomae-base-finetuned-kinetics-base-finetuned-0210

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: 0.4829
• Accuracy: 0.7627

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: 3825

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0103	0.0403	154	0.3845	0.8605
0.2216	1.0403	308	1.8650	0.6505
0.0049	2.0403	462	1.6628	0.6238
0.0019	3.0403	616	0.8134	0.6897
0.0002	4.0403	770	2.0519	0.6614
0.0158	5.0403	924	1.6577	0.6505
0.0003	6.0403	1078	1.3799	0.6473
0.0043	7.0403	1232	0.8596	0.7806
0.0006	8.0403	1386	0.9300	0.7868
2.1818	9.0403	1540	0.6332	0.7586
0.0009	10.0403	1694	2.1917	0.6458

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0