results3

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

• Loss: 0.2500
• Accuracy: 0.9661
• Recall: 0.6584
• Precision: 0.7737
• F1: 0.7114
• Roc Auc: 0.9492

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	Precision	F1	Roc Auc
0.2936	0.9988	633	0.2757	0.9337	0.6646	0.4842	0.5602	0.9290
0.2674	1.9992	1267	0.2230	0.9487	0.7391	0.5749	0.6467	0.9465
0.1575	2.9996	1901	0.2500	0.9661	0.6584	0.7737	0.7114	0.9492
0.0435	4.0	2535	0.2891	0.9613	0.7516	0.6760	0.7118	0.9476
0.0021	4.9941	3165	0.3548	0.9629	0.6770	0.7219	0.6987	0.9460

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0