videomae-base-finetuned-kinetics-0407_final_45sec_org

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: 0.4184
• Accuracy: 0.8878

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: 63500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2823	0.0100	636	0.2852	0.9142
0.2102	1.0100	1272	0.6094	0.8533
0.0022	2.0100	1908	0.2801	0.9077
0.2891	3.0100	2544	0.1889	0.9465
0.4101	4.0100	3180	0.1959	0.9299
0.1031	5.0100	3816	0.3618	0.8930
0.0792	6.0100	4452	0.3866	0.8939
0.0043	7.0100	5088	0.3914	0.8976
0.0075	8.0100	5724	0.3756	0.9105
0.0004	9.0100	6360	0.2793	0.9290
0.0033	10.0100	6996	0.3058	0.9290
0.2739	11.0100	7632	0.2605	0.9382
0.0009	12.0100	8268	0.3172	0.9188
0.5263	13.0100	8904	0.6232	0.8662
0.0001	14.0100	9540	0.5841	0.8921
0.0004	15.0100	10176	0.3614	0.9114
0.4972	16.0100	10812	0.4387	0.9087
0.0007	17.0100	11448	0.3396	0.9317
0.0001	18.0100	12084	0.4050	0.9161

Framework versions

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