metadata
license: mit
tags:
- generated_from_trainer
datasets:
- lex_glue
model-index:
- name: ECHR_test_2
results: []
ECHR_test_2
This model is a fine-tuned version of prajjwal1/bert-tiny on the lex_glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.2669
- Macro-f1: 0.4193
- Micro-f1: 0.5573
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: 0.001
- 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: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro-f1 | Micro-f1 |
|---|---|---|---|---|---|
| 0.2127 | 0.44 | 500 | 0.3104 | 0.3108 | 0.4418 |
| 0.1808 | 0.89 | 1000 | 0.2961 | 0.3452 | 0.5022 |
| 0.1762 | 1.33 | 1500 | 0.2807 | 0.3945 | 0.5273 |
| 0.16 | 1.78 | 2000 | 0.2669 | 0.4193 | 0.5573 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1