videomae-small-finetuned-kinetics-finetuned-judo

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5617
• Accuracy: 0.7838

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: 16
• eval_batch_size: 16
• 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: 380

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8417	0.0526	20	0.9854	0.5135
0.6987	1.0526	40	0.9204	0.5135
0.6783	2.0526	60	0.8633	0.5135
0.4297	3.0526	80	0.8381	0.5135
0.4407	4.0526	100	0.7284	0.6216
0.3272	5.0526	120	0.7143	0.7568
0.2406	6.0526	140	0.7214	0.7027
0.2543	7.0526	160	0.4753	0.7838
0.1978	8.0526	180	0.7151	0.7568
0.124	9.0526	200	0.5700	0.7297
0.1031	10.0526	220	0.6495	0.7027
0.0994	11.0526	240	0.4222	0.8378
0.0674	12.0526	260	0.6267	0.7568
0.0462	13.0526	280	0.6330	0.7568
0.0419	14.0526	300	0.5131	0.8108
0.0264	15.0526	320	0.4656	0.8378
0.0347	16.0526	340	0.5933	0.7838
0.0287	17.0526	360	0.5434	0.7568
0.0176	18.0526	380	0.5617	0.7838

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0