videomae-base-finetuned-kinetics-final-contest-baole1-0705

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: 0.3916
• Accuracy: 0.9087

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: 9e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 2057

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.641	0.0914	188	0.9443	0.7178
0.1614	1.0914	376	0.4797	0.8340
0.0883	2.0914	564	0.4375	0.8672
0.0963	3.0914	752	0.4745	0.8797
0.0168	4.0914	940	0.4087	0.8880
0.0011	5.0914	1128	0.3451	0.8880
0.0009	6.0914	1316	0.4118	0.8838
0.0007	7.0914	1504	0.3856	0.9046
0.0008	8.0914	1692	0.4004	0.8963
0.0006	9.0914	1880	0.4079	0.8963
0.0005	10.0860	2057	0.3916	0.9087

Framework versions

• Transformers 4.40.2
• Pytorch 1.13.1+cu117
• Datasets 2.18.0
• Tokenizers 0.19.1