videomae-base-finetuned-kinetics-0325_final

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: 0.3212
• Accuracy: 0.9102

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
0.0467	0.01	687	0.2513	0.9300
0.0084	1.01	1374	0.4972	0.8741
0.0529	2.01	2061	0.2346	0.9045
0.6585	3.01	2748	0.6326	0.8165
0.6976	4.01	3435	0.7723	0.8165
0.5654	5.01	4122	0.5031	0.8593
0.2582	6.01	4809	1.0343	0.7720
0.6946	7.01	5496	0.9468	0.7975
0.0285	8.01	6183	0.5327	0.8370
0.4778	9.01	6870	1.1722	0.7292
0.955	10.01	7557	0.6246	0.8609
0.3365	11.01	8244	0.9516	0.7728
0.0242	12.01	8931	0.6680	0.8354
0.338	13.01	9618	1.0590	0.8066
0.4445	14.01	10305	0.4882	0.8872
0.294	15.01	10992	1.1611	0.7350

Framework versions

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