metadata
license: mit
tags:
- generated_from_trainer
base_model: r1char9/rubert-base-cased-russian-sentiment
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: ru_sentiment_classification_model
results: []
ru_sentiment_classification_model
This model is a fine-tuned version of r1char9/rubert-base-cased-russian-sentiment on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6368
- Accuracy: 0.8875
- Precision: 0.8990
- Recall: 0.8875
- F1: 0.8867
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.8144 | 1.0 | 1854 | 0.7593 | 0.6630 | 0.6739 | 0.6630 | 0.6549 |
| 0.6734 | 2.0 | 3708 | 0.5688 | 0.7917 | 0.7933 | 0.7917 | 0.7899 |
| 0.5025 | 3.0 | 5562 | 0.5219 | 0.8238 | 0.8423 | 0.8238 | 0.8229 |
| 0.354 | 4.0 | 7416 | 0.4655 | 0.8912 | 0.8960 | 0.8912 | 0.8914 |
| 0.229 | 5.0 | 9270 | 0.6368 | 0.8875 | 0.8990 | 0.8875 | 0.8867 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.3.0+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1