metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: bert-finetuned-resumes-sections
results: []
bert-finetuned-resumes-sections
This model is a fine-tuned version of dbmdz/bert-base-french-europeana-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0272
- F1: 0.9625
- Roc Auc: 0.9793
- Accuracy: 0.9612
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.1714 | 1.0 | 554 | 0.0653 | 0.9301 | 0.9515 | 0.9061 |
| 0.0554 | 2.0 | 1108 | 0.0392 | 0.9489 | 0.9679 | 0.9395 |
| 0.033 | 3.0 | 1662 | 0.0318 | 0.9564 | 0.9743 | 0.9512 |
| 0.0212 | 4.0 | 2216 | 0.0295 | 0.9574 | 0.9748 | 0.9530 |
| 0.0155 | 5.0 | 2770 | 0.0282 | 0.9587 | 0.9757 | 0.9548 |
| 0.0138 | 6.0 | 3324 | 0.0282 | 0.9615 | 0.9776 | 0.9584 |
| 0.0108 | 7.0 | 3878 | 0.0272 | 0.9625 | 0.9793 | 0.9612 |
| 0.0081 | 8.0 | 4432 | 0.0284 | 0.9597 | 0.9775 | 0.9584 |
| 0.0077 | 9.0 | 4986 | 0.0267 | 0.9602 | 0.9779 | 0.9584 |
| 0.0058 | 10.0 | 5540 | 0.0281 | 0.9579 | 0.9765 | 0.9566 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1