books_text_class_roBERTa_ru_base_iliabel

This model is a fine-tuned version of DeepPavlov/xlm-roberta-large-en-ru-mnli on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2345
• Accuracy: 0.9723
• F1-score: 0.9721
• Mcc: 0.9620

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: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1-score	Mcc
0.4345	1.0	1516	0.3400	0.9190	0.9035	0.8902
0.2621	2.0	3032	0.2285	0.9563	0.9557	0.9404
0.2166	3.0	4548	0.2311	0.9661	0.9634	0.9536
0.1465	4.0	6064	0.2608	0.9606	0.9591	0.9464
0.0841	5.0	7580	0.3028	0.9581	0.9578	0.9430
0.0736	6.0	9096	0.2167	0.9735	0.9734	0.9638
0.0263	7.0	10612	0.2355	0.9738	0.9735	0.9642
0.0294	8.0	12128	0.2305	0.9711	0.9707	0.9604
0.0079	9.0	13644	0.2317	0.9726	0.9724	0.9625
0.0051	10.0	15160	0.2345	0.9723	0.9721	0.9620

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0