videomae-base-finetuned-kinetics-0325_final_lr

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: 0.5225
• Accuracy: 0.8756

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: 3e-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
0.004	0.01	687	0.6479	0.8288
0.0074	1.01	1374	0.6000	0.8486
0.0224	2.01	2061	0.3322	0.8988
0.297	3.01	2748	0.8665	0.7984
0.0564	4.01	3435	1.0001	0.8049
0.0132	5.01	4122	0.4609	0.8798
0.1194	6.01	4809	0.6630	0.8560
0.0031	7.01	5496	1.0375	0.7967
0.0372	8.01	6183	0.4677	0.9021
0.6956	9.01	6870	1.5865	0.7243
0.6914	10.01	7557	1.3021	0.7819
0.3998	11.01	8244	0.7956	0.8568
0.1396	12.01	8931	0.7252	0.8807
0.0016	13.01	9618	0.8365	0.8634
0.0072	14.01	10305	0.7298	0.8486
0.2356	15.01	10992	0.6946	0.8617
0.0304	16.01	11679	1.1166	0.8313
0.0009	17.01	12366	1.1268	0.8329
0.0003	18.01	13053	0.8779	0.8535
0.0078	19.01	13740	0.8373	0.8617
0.0024	20.01	14427	1.4814	0.8041
0.0099	21.01	15114	1.2040	0.8412
0.0188	22.01	15801	1.1281	0.8428
0.0001	23.01	16488	1.0009	0.8601

Framework versions

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