metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: media-bias-ukraine-dataset-all-removed
results: []
media-bias-ukraine-dataset-all-removed
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1717
- F1: 0.8014
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.3845 | 1.0 | 114 | 0.2296 | 0.6101 |
| 0.1412 | 2.0 | 228 | 0.1759 | 0.7486 |
| 0.0215 | 3.0 | 342 | 0.2275 | 0.7439 |
| 0.0506 | 4.0 | 456 | 0.2064 | 0.7651 |
| 0.0366 | 5.0 | 570 | 0.1717 | 0.8014 |
| 0.2428 | 6.0 | 684 | 0.1955 | 0.7878 |
| 0.005 | 7.0 | 798 | 0.2297 | 0.7839 |
| 0.003 | 8.0 | 912 | 0.2428 | 0.8005 |
| 0.0037 | 9.0 | 1026 | 0.2577 | 0.7884 |
| 0.0099 | 10.0 | 1140 | 0.2641 | 0.7957 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1
- Datasets 2.10.1
- Tokenizers 0.13.2