|
--- |
|
license: mit |
|
base_model: roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: our_data |
|
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. --> |
|
|
|
# our_data |
|
|
|
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.4250 |
|
- Precision: 0.4759 |
|
- Recall: 0.5476 |
|
- F1: 0.5092 |
|
- Accuracy: 0.7455 |
|
|
|
## 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: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 1.8353 | 0.4 | 500 | 1.6175 | 0.1212 | 0.1217 | 0.1215 | 0.5907 | |
|
| 1.4071 | 0.81 | 1000 | 1.3137 | 0.2618 | 0.3228 | 0.2891 | 0.6518 | |
|
| 1.1532 | 1.21 | 1500 | 1.2950 | 0.3154 | 0.3558 | 0.3344 | 0.6739 | |
|
| 0.9969 | 1.61 | 2000 | 1.1882 | 0.3266 | 0.4034 | 0.3609 | 0.6783 | |
|
| 0.922 | 2.01 | 2500 | 1.2653 | 0.3471 | 0.3995 | 0.3715 | 0.6873 | |
|
| 0.739 | 2.42 | 3000 | 1.1592 | 0.3538 | 0.4339 | 0.3898 | 0.7034 | |
|
| 0.6866 | 2.82 | 3500 | 1.2015 | 0.3521 | 0.4299 | 0.3871 | 0.7017 | |
|
| 0.5554 | 3.22 | 4000 | 1.2555 | 0.4398 | 0.4643 | 0.4517 | 0.7329 | |
|
| 0.5009 | 3.63 | 4500 | 1.2871 | 0.4098 | 0.4868 | 0.4450 | 0.7230 | |
|
| 0.5117 | 4.03 | 5000 | 1.2482 | 0.4030 | 0.4974 | 0.4452 | 0.7279 | |
|
| 0.3771 | 4.43 | 5500 | 1.3005 | 0.4300 | 0.4960 | 0.4607 | 0.7261 | |
|
| 0.4357 | 4.83 | 6000 | 1.2412 | 0.4516 | 0.5251 | 0.4856 | 0.7395 | |
|
| 0.3151 | 5.24 | 6500 | 1.3410 | 0.4423 | 0.5225 | 0.4791 | 0.7333 | |
|
| 0.3219 | 5.64 | 7000 | 1.2903 | 0.425 | 0.5172 | 0.4666 | 0.7366 | |
|
| 0.3405 | 6.04 | 7500 | 1.3366 | 0.4470 | 0.5304 | 0.4852 | 0.7471 | |
|
| 0.2856 | 6.45 | 8000 | 1.3243 | 0.4415 | 0.5344 | 0.4835 | 0.7474 | |
|
| 0.2723 | 6.85 | 8500 | 1.3962 | 0.4540 | 0.5291 | 0.4887 | 0.7398 | |
|
| 0.2307 | 7.25 | 9000 | 1.4783 | 0.4671 | 0.5357 | 0.4991 | 0.7440 | |
|
| 0.2484 | 7.66 | 9500 | 1.4250 | 0.4759 | 0.5476 | 0.5092 | 0.7455 | |
|
| 0.2361 | 8.06 | 10000 | 1.4695 | 0.4700 | 0.5384 | 0.5018 | 0.7518 | |
|
| 0.186 | 8.46 | 10500 | 1.5283 | 0.4587 | 0.5516 | 0.5009 | 0.7520 | |
|
| 0.2188 | 8.86 | 11000 | 1.4357 | 0.4478 | 0.5450 | 0.4916 | 0.7471 | |
|
| 0.2072 | 9.27 | 11500 | 1.4810 | 0.4770 | 0.5357 | 0.5047 | 0.7527 | |
|
| 0.1817 | 9.67 | 12000 | 1.5041 | 0.4719 | 0.5450 | 0.5058 | 0.7532 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0.dev0 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
