videomae-base-finetuned-kinetics-finetuned-right-hand-conflab-v3

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: 1.3489
• Accuracy: 0.5813

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 936

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.9096	0.1261	118	1.6888	0.3333
1.628	1.1261	236	1.6285	0.4118
1.3656	2.1261	354	1.3947	0.5147
1.1498	3.1261	472	1.2639	0.5735
0.6546	4.1261	590	1.2054	0.5882
0.4812	5.1261	708	1.2695	0.5490
0.4631	6.1261	826	1.2095	0.6078
0.2311	7.1175	936	1.1960	0.5980

Framework versions

• Transformers 4.41.0
• Pytorch 1.12.0+cu116
• Datasets 2.19.1
• Tokenizers 0.19.1