metadata
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: distilr-lr1e05-wd0.05-bs32
results: []
distilr-lr1e05-wd0.05-bs32
This model is a fine-tuned version of distilroberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2743
- Rmse: 0.5237
- Mse: 0.2743
- Mae: 0.4135
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: 1e-05
- train_batch_size: 256
- eval_batch_size: 256
- 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 |
|---|---|---|---|---|---|---|
| 0.2775 | 1.0 | 623 | 0.2735 | 0.5229 | 0.2735 | 0.4179 |
| 0.2738 | 2.0 | 1246 | 0.2727 | 0.5222 | 0.2727 | 0.4126 |
| 0.2722 | 3.0 | 1869 | 0.2727 | 0.5222 | 0.2727 | 0.4165 |
| 0.2702 | 4.0 | 2492 | 0.2754 | 0.5248 | 0.2754 | 0.3997 |
| 0.2684 | 5.0 | 3115 | 0.2765 | 0.5259 | 0.2765 | 0.4229 |
| 0.2668 | 6.0 | 3738 | 0.2743 | 0.5237 | 0.2743 | 0.4135 |
Framework versions
- Transformers 4.19.0.dev0
- Pytorch 1.9.0+cu111
- Datasets 2.4.0
- Tokenizers 0.12.1