|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: bert-nlp-project-ft-imdb |
|
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. --> |
|
|
|
# bert-nlp-project-ft-imdb |
|
|
|
This model is a fine-tuned version of [jestemleon/bert-nlp-project-imdb](https://huggingface.co/jestemleon/bert-nlp-project-imdb) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2423 |
|
- Accuracy: 0.948 |
|
- F1: 0.9472 |
|
|
|
## 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: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:| |
|
| 0.2653 | 0.38 | 750 | 0.1871 | 0.9325 | 0.9325 | |
|
| 0.203 | 0.75 | 1500 | 0.1799 | 0.9403 | 0.9371 | |
|
| 0.1703 | 1.12 | 2250 | 0.1893 | 0.9473 | 0.9467 | |
|
| 0.1206 | 1.5 | 3000 | 0.2015 | 0.9475 | 0.9467 | |
|
| 0.1085 | 1.88 | 3750 | 0.2163 | 0.9465 | 0.9464 | |
|
| 0.0738 | 2.25 | 4500 | 0.2520 | 0.9457 | 0.9454 | |
|
| 0.0564 | 2.62 | 5250 | 0.2491 | 0.9467 | 0.9454 | |
|
| 0.0512 | 3.0 | 6000 | 0.2423 | 0.948 | 0.9472 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.25.1 |
|
- Pytorch 1.13.0+cu116 |
|
- Datasets 2.7.1 |
|
- Tokenizers 0.13.2 |
|
|
