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: 0.6582
• Accuracy: 0.8824

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
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 488

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.198	0.1270	62	1.4080	0.5532
0.8081	1.1270	124	1.5203	0.6596
0.3909	2.1270	186	1.2482	0.7021
0.2733	3.1270	248	1.1558	0.7447
0.0197	4.1270	310	0.9275	0.7660
0.0911	5.1270	372	1.1365	0.7660
0.0514	6.1270	434	0.7995	0.7872
0.0023	7.1107	488	0.7915	0.7872

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1