metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: output
results: []
output
This model is a fine-tuned version of HooshvareLab/bert-fa-zwnj-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.9434
- Accuracy: 0.6684
- F1: 0.6608
- F1 Micro: 0.6684
- F1 Macro: 0.5384
- Precision: 0.6585
- Recall: 0.6684
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: 0.0001
- 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
- lr_scheduler_warmup_steps: 50
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | F1 Micro | F1 Macro | Precision | Recall |
|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 96 | 1.1053 | 0.5789 | 0.5598 | 0.5789 | 0.4428 | 0.5993 | 0.5789 |
| No log | 2.0 | 192 | 1.1474 | 0.5737 | 0.5831 | 0.5737 | 0.4693 | 0.6383 | 0.5737 |
| No log | 3.0 | 288 | 1.0758 | 0.6316 | 0.6016 | 0.6316 | 0.4850 | 0.6356 | 0.6316 |
| No log | 4.0 | 384 | 1.2156 | 0.6474 | 0.6454 | 0.6474 | 0.5224 | 0.6489 | 0.6474 |
| No log | 5.0 | 480 | 1.5757 | 0.6316 | 0.6276 | 0.6316 | 0.5099 | 0.6293 | 0.6316 |
| 0.5922 | 6.0 | 576 | 1.7277 | 0.6579 | 0.6542 | 0.6579 | 0.5338 | 0.6520 | 0.6579 |
| 0.5922 | 7.0 | 672 | 1.9130 | 0.6789 | 0.6736 | 0.6789 | 0.5482 | 0.6749 | 0.6789 |
| 0.5922 | 8.0 | 768 | 1.9434 | 0.6684 | 0.6608 | 0.6684 | 0.5384 | 0.6585 | 0.6684 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Tokenizers 0.13.3