metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: finetuned_distilbert_fa_zwnj_base_ner
results: []
finetuned_distilbert_fa_zwnj_base_ner
This model is a fine-tuned version of HooshvareLab/distilbert-fa-zwnj-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0296
- Precision: 0.9199
- Recall: 0.9378
- F1: 0.9288
- Accuracy: 0.9923
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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1443 | 1.0 | 1820 | 0.0708 | 0.7875 | 0.7912 | 0.7894 | 0.9771 |
| 0.0551 | 2.0 | 3640 | 0.0443 | 0.8602 | 0.8918 | 0.8757 | 0.9863 |
| 0.0289 | 3.0 | 5460 | 0.0343 | 0.8914 | 0.9216 | 0.9062 | 0.9899 |
| 0.0171 | 4.0 | 7280 | 0.0303 | 0.9142 | 0.9333 | 0.9236 | 0.9918 |
| 0.0117 | 5.0 | 9100 | 0.0296 | 0.9199 | 0.9378 | 0.9288 | 0.9923 |
Framework versions
- Transformers 4.21.2
- Pytorch 1.12.1+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1