mahsharyahan/DeepPavlov_rubert_conv_RU

DeepPavlov_rubert_RU

This model is a fine-tuned version of DeepPavlov/rubert-base-cased-conversational on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4864
• Accuracy: 0.8214
• F1: 0.8718
• Precision: 0.8293
• Recall: 0.9189

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• 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: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	25	0.5777	0.7143	0.8222	0.6981	1.0
No log	2.0	50	0.5717	0.6786	0.7955	0.6863	0.9459
No log	3.0	75	0.5039	0.7857	0.85	0.7907	0.9189
No log	4.0	100	0.5014	0.8036	0.8608	0.8095	0.9189
No log	5.0	125	0.4864	0.8214	0.8718	0.8293	0.9189

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0