metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: idp-headers
results: []
idp-headers
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.6714
- F1: 0.4823
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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 1.7995 | 1.0 | 5 | 1.8557 | 0.1629 |
| 1.7125 | 2.0 | 10 | 1.7832 | 0.1759 |
| 1.6381 | 3.0 | 15 | 1.7243 | 0.4698 |
| 1.5746 | 4.0 | 20 | 1.6857 | 0.4823 |
| 1.5354 | 5.0 | 25 | 1.6714 | 0.4823 |
Framework versions
- Transformers 4.21.3
- Pytorch 1.12.1+cu113
- Tokenizers 0.12.1