RELEXset-Predictor / README.md
paola-md's picture
Update README.md
1de5567 verified
---
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: recipe-lr5e05-wd0.02-bs32
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. -->
# recipe-lr5e05-wd0.02-bs32
This model is a fine-tuned version of [paola-md/recipe-distilroberta-Is](https://huggingface.co/paola-md/recipe-distilroberta-Is) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2738
- Rmse: 0.5232
- Mse: 0.2738
- Mae: 0.4117
## 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: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Rmse | Mse | Mae |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|
| 0.2788 | 1.0 | 1245 | 0.2766 | 0.5259 | 0.2766 | 0.4212 |
| 0.2757 | 2.0 | 2490 | 0.2777 | 0.5270 | 0.2777 | 0.4271 |
| 0.2741 | 3.0 | 3735 | 0.2745 | 0.5239 | 0.2745 | 0.4202 |
| 0.2725 | 4.0 | 4980 | 0.2760 | 0.5254 | 0.2760 | 0.4030 |
| 0.2711 | 5.0 | 6225 | 0.2752 | 0.5246 | 0.2752 | 0.4186 |
| 0.2692 | 6.0 | 7470 | 0.2738 | 0.5232 | 0.2738 | 0.4117 |
### Test results
Using the model saved for 4 epochs, we achived the following results on the test set:
```
{
"test_mae": 0.39440343575349946,
"test_runtime": 22.6921,
"test_samples_per_second": 866.867,
"test_steps_per_second": 6.787
}
```
### Framework versions
- Transformers 4.19.0.dev0
- Pytorch 1.9.0+cu111
- Datasets 2.4.0
- Tokenizers 0.12.1