videomae-base-finetuned-kinetics-0409_final_5sec_org_ab7_val_inside_train

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: 0.3255
• Accuracy: 0.9138

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: 67100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4404	0.0100	672	0.2641	0.8958
0.0506	1.0100	1344	0.2971	0.9045
0.0077	2.0100	2016	0.8203	0.8293
0.016	3.0100	2688	0.4447	0.8958
0.0012	4.0100	3360	0.5228	0.8622
0.0003	5.0100	4032	0.5333	0.8731
0.0019	6.0100	4704	0.5615	0.8786
0.0669	7.0100	5376	0.3206	0.9162
0.0056	8.0100	6048	0.5627	0.8849
0.0003	9.0100	6720	0.6655	0.8567
0.0063	10.0100	7392	0.6566	0.8786
0.0003	11.0100	8064	0.5058	0.8778
0.0005	12.0100	8736	0.4329	0.9045
0.0005	13.0100	9408	0.4837	0.8943
0.0182	14.0100	10080	0.6702	0.8692
0.0001	15.0100	10752	0.7277	0.8583
0.0001	16.0100	11424	0.6110	0.8763
0.0001	17.0100	12096	0.5027	0.9146
0.0006	18.0100	12768	0.4604	0.9068
0.0138	19.0100	13440	0.4703	0.9123
0.0	20.0100	14112	0.4712	0.9068
0.5385	21.0100	14784	0.5793	0.9021
0.0001	22.0100	15456	0.4995	0.9162

Framework versions

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