videomae-base-finetuned-kinetics-finetuned-right-hand-conflab-v4

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: 1.9989
• Accuracy: 0.5764

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: 1287

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1446	0.0917	118	1.7563	0.5931
0.1123	1.0917	236	1.7033	0.5833
0.3621	2.0917	354	2.0058	0.5686
0.0979	3.0917	472	1.9469	0.5784
0.2599	4.0917	590	1.9703	0.6029
0.0292	5.0917	708	1.9382	0.6275
0.0055	6.0917	826	2.0823	0.6176
0.0056	7.0917	944	2.0926	0.5931
0.0016	8.0917	1062	2.1117	0.6176
0.0013	9.0917	1180	2.1569	0.6225
0.2274	10.0831	1287	2.0956	0.6078

Framework versions

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