ruBert-large-topic_classification

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

• Loss: 0.7900
• Precision: 0.8793
• Recall: 0.8646
• F1: 0.8688
• Accuracy: 0.8824

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: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	44	0.6327	0.8534	0.7883	0.8046	0.8186
No log	2.0	88	0.4897	0.8847	0.8401	0.8548	0.8676
No log	3.0	132	0.5957	0.8732	0.8617	0.8638	0.8676
No log	4.0	176	0.6598	0.8808	0.8658	0.8700	0.8824
No log	5.0	220	0.7086	0.8705	0.8589	0.8625	0.8775
No log	6.0	264	0.7445	0.8793	0.8646	0.8688	0.8824
No log	7.0	308	0.7661	0.8793	0.8646	0.8688	0.8824
No log	8.0	352	0.7795	0.8793	0.8646	0.8688	0.8824
No log	9.0	396	0.7870	0.8793	0.8646	0.8688	0.8824
No log	10.0	440	0.7900	0.8793	0.8646	0.8688	0.8824

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1