videomae-base-finetuned-ucf101-subset

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3402
• Accuracy: 0.9054

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: 3530

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2221	0.0989	349	1.0476	0.6204
0.7062	1.0989	698	0.7049	0.7679
0.626	2.0989	1047	0.4938	0.8394
0.4414	3.0989	1396	0.5213	0.8365
0.4938	4.0989	1745	0.5057	0.8248
0.4516	5.0989	2094	0.4209	0.8730
0.2555	6.0989	2443	0.4210	0.8701
0.3595	7.0989	2792	0.3525	0.8993
0.2301	8.0989	3141	0.3473	0.9036
0.1837	9.0989	3490	0.3532	0.8949
0.3946	10.0113	3530	0.3517	0.8964

Framework versions

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