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: 0.4251
• Accuracy: 0.9

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 90

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3225	0.1111	10	2.1767	0.3286
2.0787	1.1111	20	1.8950	0.4571
1.5825	2.1111	30	1.4276	0.6429
1.0043	3.1111	40	0.9962	0.6714
0.6267	4.1111	50	0.7575	0.8286
0.4496	5.1111	60	0.6497	0.8429
0.3364	6.1111	70	0.5238	0.8143
0.244	7.1111	80	0.4385	0.8429
0.1867	8.1111	90	0.4251	0.9

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1