| --- |
| base_model: s-nlp/russian_toxicity_classifier |
| tags: |
| - generated_from_trainer |
| metrics: |
| - precision |
| - recall |
| - f1 |
| - accuracy |
| model-index: |
| - name: tg_comments_model |
| results: [] |
| --- |
| |
| <!-- This model card has been generated automatically according to the information the Trainer had access to. You |
| should probably proofread and complete it, then remove this comment. --> |
|
|
| # tg_comments_model |
|
|
| This model is a fine-tuned version of [s-nlp/russian_toxicity_classifier](https://huggingface.co/s-nlp/russian_toxicity_classifier) on the None dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.0519 |
| - Precision: 0.9762 |
| - Recall: 0.9856 |
| - F1: 0.9809 |
| - Accuracy: 0.9817 |
|
|
| ## 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: 3e-05 |
| - train_batch_size: 64 |
| - eval_batch_size: 32 |
| - seed: 42 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: linear |
| - lr_scheduler_warmup_steps: 10 |
| - num_epochs: 1 |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
| |:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
| | 0.075 | 0.2239 | 300 | 0.0591 | 0.9833 | 0.9731 | 0.9781 | 0.9793 | |
| | 0.0627 | 0.4478 | 600 | 0.0567 | 0.9749 | 0.9843 | 0.9796 | 0.9805 | |
| | 0.0612 | 0.6716 | 900 | 0.0537 | 0.9795 | 0.9821 | 0.9808 | 0.9817 | |
| | 0.0633 | 0.8955 | 1200 | 0.0519 | 0.9762 | 0.9856 | 0.9809 | 0.9817 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 4.41.2 |
| - Pytorch 2.3.0+cu121 |
| - Datasets 2.20.0 |
| - Tokenizers 0.19.1 |
| |
