profesor_MViT_B_RLVS

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

• Loss: 0.0380
• Accuracy: 0.9908
• F1: 0.9908
• Precision: 0.9908
• Recall: 0.9908
• Roc Auc: 0.9992

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: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Use 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_steps: 240
• training_steps: 2400
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Roc Auc
0.427	2.0333	240	0.2509	0.9661	0.9660	0.9671	0.9661	0.9961
0.1551	5.0333	480	0.1313	0.9687	0.9687	0.9695	0.9687	0.9974
0.0783	8.0333	720	0.1090	0.9713	0.9713	0.9719	0.9713	0.9974
0.0547	11.0333	960	0.0647	0.9843	0.9843	0.9843	0.9843	0.9974
0.0379	14.0333	1200	0.0472	0.9896	0.9896	0.9896	0.9896	0.9987
0.0337	17.0333	1440	0.0552	0.9869	0.9869	0.9870	0.9869	0.9986
0.0299	20.0333	1680	0.0786	0.9843	0.9843	0.9844	0.9843	0.9992
0.0308	23.0333	1920	0.0753	0.9817	0.9817	0.9818	0.9817	0.9995

Framework versions

• Transformers 4.46.1
• Pytorch 2.0.1+cu118
• Datasets 3.0.2
• Tokenizers 0.20.1