metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: camembert-ner-lr10e6
results: []
camembert-ner-lr10e6
This model is a fine-tuned version of Jean-Baptiste/camembert-ner on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4541
- Overall Precision: 0.3799
- Overall Recall: 0.4559
- Overall F1: 0.4144
- Overall Accuracy: 0.9163
- Humanprod F1: 0.0
- Loc F1: 0.3266
- Org F1: 0.0323
- Per F1: 0.5821
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-06
- 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: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Humanprod F1 | Loc F1 | Org F1 | Per F1 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.9048 | 1.0 | 613 | 0.4945 | 0.3748 | 0.4036 | 0.3887 | 0.9151 | 0.0 | 0.3447 | 0.0318 | 0.5020 |
| 0.5001 | 2.0 | 1226 | 0.4541 | 0.3799 | 0.4559 | 0.4144 | 0.9163 | 0.0 | 0.3266 | 0.0323 | 0.5821 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.7.1+cpu
- Datasets 2.7.1
- Tokenizers 0.13.2