vivit-b-16x2-kinetics400-finetuned-ucf101-subset

This model is a fine-tuned version of google/vivit-b-16x2-kinetics400 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6545
• Accuracy: 0.8339

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 7060

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1788	0.0494	349	1.1337	0.5608
0.93	1.0494	698	1.0407	0.6386
0.7623	2.0494	1047	0.8954	0.6775
0.6178	3.0494	1396	0.8011	0.7050
0.4934	4.0494	1745	0.8328	0.7034
0.5447	5.0494	2094	0.7323	0.7731
0.4106	6.0494	2443	0.6595	0.7763
0.4948	7.0494	2792	0.6986	0.7780
0.3257	8.0494	3141	0.5867	0.8152
0.2109	9.0494	3490	0.6624	0.8039
0.3482	10.0494	3839	0.6958	0.8104
0.4104	11.0494	4188	0.5619	0.8331
0.2176	12.0494	4537	0.5727	0.8217
0.1634	13.0494	4886	0.4984	0.8606
0.1256	14.0494	5235	0.5515	0.8444
0.188	15.0494	5584	0.5470	0.8428
0.2174	16.0494	5933	0.4847	0.8590
0.1619	17.0494	6282	0.5030	0.8784
0.3518	18.0494	6631	0.4783	0.8784
0.1533	19.0494	6980	0.5133	0.8703
0.101	20.0113	7060	0.5090	0.8703

Framework versions

• Transformers 4.44.1
• Pytorch 2.0.1+cu117
• Datasets 2.21.0
• Tokenizers 0.19.1