metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- matthews_correlation
model-index:
- name: distilbert-base-uncased-cola
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: glue
type: glue
args: cola
metrics:
- name: Matthews Correlation
type: matthews_correlation
value: 0.5474713423103301
distilbert-base-uncased-cola
This model is a fine-tuned version of sciarrilli/distilbert-base-uncased-cola on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 1.2934
- Matthews Correlation: 0.5475
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
|---|---|---|---|---|
| 0.3525 | 1.0 | 535 | 0.5328 | 0.4268 |
| 0.2941 | 2.0 | 1070 | 0.6235 | 0.4805 |
| 0.2111 | 3.0 | 1605 | 0.6743 | 0.4982 |
| 0.1607 | 4.0 | 2140 | 0.7463 | 0.5172 |
| 0.1487 | 5.0 | 2675 | 0.9553 | 0.4935 |
| 0.102 | 6.0 | 3210 | 1.0206 | 0.5121 |
| 0.0878 | 7.0 | 3745 | 1.1293 | 0.5347 |
| 0.0605 | 8.0 | 4280 | 1.1958 | 0.5422 |
| 0.0418 | 9.0 | 4815 | 1.2673 | 0.5449 |
| 0.0378 | 10.0 | 5350 | 1.2934 | 0.5475 |
Framework versions
- Transformers 4.12.3
- Pytorch 1.10.0+cu102
- Datasets 1.15.1
- Tokenizers 0.10.3