metadata
base_model: HooshvareLab/bert-base-parsbert-uncased
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: Persian-Text-Sentiment-Bert-V1
results: []
language:
- fa
Persian-Text-Sentiment-Bert-V1
This model is a fine-tuned version of HooshvareLab/bert-base-parsbert-uncased on a custom dataset. It achieves the following results on the evaluation set:
- Loss: 0.3265
- Precision: 0.8727
- Recall: 0.8716
- F1-score: 0.8715
- Accuracy: 0.8716
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: 16
- eval_batch_size: 16
- 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-score | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.3097 | 1.0 | 3491 | 0.3265 | 0.8727 | 0.8716 | 0.8715 | 0.8716 |
| 0.2686 | 2.0 | 6982 | 0.3602 | 0.8785 | 0.8758 | 0.8756 | 0.8758 |
| 0.2137 | 3.0 | 10473 | 0.3828 | 0.8759 | 0.8724 | 0.8721 | 0.8724 |
| 0.1823 | 4.0 | 13964 | 0.5545 | 0.8637 | 0.8636 | 0.8636 | 0.8636 |
| 0.1346 | 5.0 | 17455 | 0.6295 | 0.8572 | 0.8566 | 0.8566 | 0.8566 |
| 0.1001 | 6.0 | 20946 | 0.8501 | 0.8606 | 0.8604 | 0.8604 | 0.8604 |
| 0.071 | 7.0 | 24437 | 1.0192 | 0.8596 | 0.8594 | 0.8594 | 0.8594 |
| 0.0604 | 8.0 | 27928 | 1.0449 | 0.8553 | 0.8553 | 0.8553 | 0.8553 |
| 0.0312 | 9.0 | 31419 | 1.1677 | 0.8598 | 0.8598 | 0.8598 | 0.8598 |
| 0.022 | 10.0 | 34910 | 1.2128 | 0.8593 | 0.8591 | 0.8591 | 0.8591 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3