ALL_RGBCROP_ori16F-8B16F-GWlr-cosine

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

• Loss: 0.3658
• Accuracy: 0.8623

Best Checkpoint : 240

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 1152

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6775	0.0417	48	0.7147	0.4695
0.5877	1.0417	96	0.6383	0.6220
0.3375	2.0417	144	0.5176	0.7317
0.219	3.0417	192	0.4915	0.7805
0.0698	4.0417	240	0.5611	0.8110
0.0587	5.0417	288	0.6506	0.7927
0.0194	6.0417	336	0.7638	0.7988
0.0029	7.0417	384	0.9139	0.7805
0.0023	8.0417	432	0.9306	0.7988
0.0033	9.0417	480	0.9203	0.7988

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1