|
--- |
|
license: mit |
|
base_model: camembert-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: camembert_classification_tools_qlora_fr |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# camembert_classification_tools_qlora_fr |
|
|
|
This model is a fine-tuned version of [camembert-base](https://huggingface.co/camembert-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7280 |
|
- Accuracy: 0.85 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 24 |
|
- eval_batch_size: 192 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 60 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 1.0 | 5 | 2.0968 | 0.075 | |
|
| No log | 2.0 | 10 | 2.1103 | 0.075 | |
|
| No log | 3.0 | 15 | 2.1119 | 0.075 | |
|
| No log | 4.0 | 20 | 2.1045 | 0.1 | |
|
| No log | 5.0 | 25 | 2.0948 | 0.125 | |
|
| No log | 6.0 | 30 | 2.0586 | 0.125 | |
|
| No log | 7.0 | 35 | 2.0199 | 0.2 | |
|
| No log | 8.0 | 40 | 1.9633 | 0.25 | |
|
| No log | 9.0 | 45 | 1.9075 | 0.35 | |
|
| No log | 10.0 | 50 | 1.8445 | 0.5 | |
|
| No log | 11.0 | 55 | 1.7872 | 0.55 | |
|
| No log | 12.0 | 60 | 1.7288 | 0.6 | |
|
| No log | 13.0 | 65 | 1.6744 | 0.625 | |
|
| No log | 14.0 | 70 | 1.6192 | 0.65 | |
|
| No log | 15.0 | 75 | 1.5612 | 0.65 | |
|
| No log | 16.0 | 80 | 1.5041 | 0.65 | |
|
| No log | 17.0 | 85 | 1.4466 | 0.7 | |
|
| No log | 18.0 | 90 | 1.3910 | 0.675 | |
|
| No log | 19.0 | 95 | 1.3369 | 0.7 | |
|
| No log | 20.0 | 100 | 1.2929 | 0.725 | |
|
| No log | 21.0 | 105 | 1.2470 | 0.725 | |
|
| No log | 22.0 | 110 | 1.2048 | 0.725 | |
|
| No log | 23.0 | 115 | 1.1597 | 0.725 | |
|
| No log | 24.0 | 120 | 1.1148 | 0.775 | |
|
| No log | 25.0 | 125 | 1.0787 | 0.775 | |
|
| No log | 26.0 | 130 | 1.0485 | 0.775 | |
|
| No log | 27.0 | 135 | 1.0222 | 0.75 | |
|
| No log | 28.0 | 140 | 0.9954 | 0.8 | |
|
| No log | 29.0 | 145 | 0.9714 | 0.85 | |
|
| No log | 30.0 | 150 | 0.9442 | 0.825 | |
|
| No log | 31.0 | 155 | 0.9201 | 0.85 | |
|
| No log | 32.0 | 160 | 0.9032 | 0.85 | |
|
| No log | 33.0 | 165 | 0.8843 | 0.85 | |
|
| No log | 34.0 | 170 | 0.8739 | 0.85 | |
|
| No log | 35.0 | 175 | 0.8527 | 0.85 | |
|
| No log | 36.0 | 180 | 0.8312 | 0.85 | |
|
| No log | 37.0 | 185 | 0.8193 | 0.85 | |
|
| No log | 38.0 | 190 | 0.8079 | 0.85 | |
|
| No log | 39.0 | 195 | 0.8015 | 0.85 | |
|
| No log | 40.0 | 200 | 0.7962 | 0.85 | |
|
| No log | 41.0 | 205 | 0.7958 | 0.85 | |
|
| No log | 42.0 | 210 | 0.7846 | 0.85 | |
|
| No log | 43.0 | 215 | 0.7652 | 0.85 | |
|
| No log | 44.0 | 220 | 0.7536 | 0.85 | |
|
| No log | 45.0 | 225 | 0.7451 | 0.85 | |
|
| No log | 46.0 | 230 | 0.7377 | 0.85 | |
|
| No log | 47.0 | 235 | 0.7327 | 0.85 | |
|
| No log | 48.0 | 240 | 0.7310 | 0.85 | |
|
| No log | 49.0 | 245 | 0.7312 | 0.85 | |
|
| No log | 50.0 | 250 | 0.7280 | 0.85 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.0 |
|
- Pytorch 2.0.1+cu117 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.14.1 |
|
|
