|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: BERiT_2000 |
|
|
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. --> |
|
|
|
|
|
# BERiT_2000 |
|
|
|
|
|
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: 6.7293 |
|
|
|
|
|
## 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: 5e-05 |
|
|
- train_batch_size: 8 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 3 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:----:|:---------------:| |
|
|
| 6.9294 | 0.19 | 500 | 6.8136 | |
|
|
| 6.7692 | 0.39 | 1000 | 6.8006 | |
|
|
| 6.7567 | 0.58 | 1500 | 6.7770 | |
|
|
| 6.746 | 0.77 | 2000 | 6.7414 | |
|
|
| 6.7577 | 0.97 | 2500 | 6.7333 | |
|
|
| 6.7295 | 1.16 | 3000 | 6.7405 | |
|
|
| 6.7635 | 1.36 | 3500 | 6.7272 | |
|
|
| 6.7715 | 1.55 | 4000 | 6.7114 | |
|
|
| 6.7348 | 1.74 | 4500 | 6.7275 | |
|
|
| 6.719 | 1.94 | 5000 | 6.7322 | |
|
|
| 6.7427 | 2.13 | 5500 | 6.7242 | |
|
|
| 6.7136 | 2.32 | 6000 | 6.6852 | |
|
|
| 6.719 | 2.52 | 6500 | 6.7430 | |
|
|
| 6.7229 | 2.71 | 7000 | 6.7331 | |
|
|
| 6.7166 | 2.9 | 7500 | 6.7293 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.24.0 |
|
|
- Pytorch 1.12.1+cu113 |
|
|
- Datasets 2.6.1 |
|
|
- Tokenizers 0.13.2 |
|
|
|
