camembert_squadFR_question_answering_tools_qlora_fr
This model is a fine-tuned version of etalab-ia/camembert-base-squadFR-fquad-piaf on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3994
- Learning Rate: 0.0
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 | Rate |
|---|---|---|---|---|
| No log | 1.0 | 7 | 3.8843 | 0.0001 |
| No log | 2.0 | 14 | 2.5855 | 0.0001 |
| No log | 3.0 | 21 | 1.9334 | 0.0001 |
| No log | 4.0 | 28 | 1.5336 | 0.0001 |
| No log | 5.0 | 35 | 1.2781 | 0.0001 |
| No log | 6.0 | 42 | 1.0961 | 9e-05 |
| No log | 7.0 | 49 | 1.0501 | 0.0001 |
| No log | 8.0 | 56 | 1.0563 | 0.0001 |
| No log | 9.0 | 63 | 1.0709 | 0.0001 |
| No log | 10.0 | 70 | 1.0806 | 0.0001 |
| No log | 11.0 | 77 | 1.1127 | 0.0001 |
| No log | 12.0 | 84 | 1.0596 | 8e-05 |
| No log | 13.0 | 91 | 1.0592 | 0.0001 |
| No log | 14.0 | 98 | 1.0619 | 0.0001 |
| No log | 15.0 | 105 | 1.0277 | 0.0001 |
| No log | 16.0 | 112 | 0.9977 | 0.0001 |
| No log | 17.0 | 119 | 1.0214 | 0.0001 |
| No log | 18.0 | 126 | 1.0584 | 7e-05 |
| No log | 19.0 | 133 | 1.0866 | 0.0001 |
| No log | 20.0 | 140 | 1.1212 | 0.0001 |
| No log | 21.0 | 147 | 1.0932 | 0.0001 |
| No log | 22.0 | 154 | 1.1510 | 0.0001 |
| No log | 23.0 | 161 | 1.1591 | 0.0001 |
| No log | 24.0 | 168 | 1.1126 | 6e-05 |
| No log | 25.0 | 175 | 1.1220 | 0.0001 |
| No log | 26.0 | 182 | 1.1413 | 0.0001 |
| No log | 27.0 | 189 | 1.1519 | 0.0001 |
| No log | 28.0 | 196 | 1.1718 | 0.0001 |
| No log | 29.0 | 203 | 1.2205 | 0.0001 |
| No log | 30.0 | 210 | 1.2722 | 5e-05 |
| No log | 31.0 | 217 | 1.2453 | 0.0000 |
| No log | 32.0 | 224 | 1.2521 | 0.0000 |
| No log | 33.0 | 231 | 1.2604 | 0.0000 |
| No log | 34.0 | 238 | 1.2694 | 0.0000 |
| No log | 35.0 | 245 | 1.2688 | 0.0000 |
| No log | 36.0 | 252 | 1.2813 | 4e-05 |
| No log | 37.0 | 259 | 1.3230 | 0.0000 |
| No log | 38.0 | 266 | 1.3333 | 0.0000 |
| No log | 39.0 | 273 | 1.2986 | 0.0000 |
| No log | 40.0 | 280 | 1.3076 | 0.0000 |
| No log | 41.0 | 287 | 1.3551 | 0.0000 |
| No log | 42.0 | 294 | 1.3552 | 3e-05 |
| No log | 43.0 | 301 | 1.3435 | 0.0000 |
| No log | 44.0 | 308 | 1.3404 | 0.0000 |
| No log | 45.0 | 315 | 1.3245 | 0.0000 |
| No log | 46.0 | 322 | 1.3288 | 0.0000 |
| No log | 47.0 | 329 | 1.3455 | 0.0000 |
| No log | 48.0 | 336 | 1.3656 | 2e-05 |
| No log | 49.0 | 343 | 1.4061 | 0.0000 |
| No log | 50.0 | 350 | 1.4262 | 0.0000 |
| No log | 51.0 | 357 | 1.4366 | 0.0000 |
| No log | 52.0 | 364 | 1.4267 | 0.0000 |
| No log | 53.0 | 371 | 1.4158 | 0.0000 |
| No log | 54.0 | 378 | 1.4038 | 1e-05 |
| No log | 55.0 | 385 | 1.4013 | 0.0000 |
| No log | 56.0 | 392 | 1.4022 | 0.0000 |
| No log | 57.0 | 399 | 1.3999 | 5e-06 |
| No log | 58.0 | 406 | 1.3987 | 0.0000 |
| No log | 59.0 | 413 | 1.3994 | 0.0000 |
| No log | 60.0 | 420 | 1.3994 | 0.0 |
Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.14.1