File size: 5,044 Bytes
e5bd53e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 |
---
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: 0.7203
- Accuracy: 0.875
## 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 | 2.0791 | 0.075 |
| No log | 2.0 | 10 | 2.0909 | 0.075 |
| No log | 3.0 | 15 | 2.0903 | 0.075 |
| No log | 4.0 | 20 | 2.0790 | 0.075 |
| No log | 5.0 | 25 | 2.0606 | 0.075 |
| No log | 6.0 | 30 | 2.0206 | 0.1 |
| No log | 7.0 | 35 | 1.9780 | 0.25 |
| No log | 8.0 | 40 | 1.9250 | 0.375 |
| No log | 9.0 | 45 | 1.8724 | 0.5 |
| No log | 10.0 | 50 | 1.8129 | 0.525 |
| No log | 11.0 | 55 | 1.7570 | 0.55 |
| No log | 12.0 | 60 | 1.7009 | 0.65 |
| No log | 13.0 | 65 | 1.6472 | 0.625 |
| No log | 14.0 | 70 | 1.5928 | 0.675 |
| No log | 15.0 | 75 | 1.5434 | 0.7 |
| No log | 16.0 | 80 | 1.4880 | 0.675 |
| No log | 17.0 | 85 | 1.4333 | 0.7 |
| No log | 18.0 | 90 | 1.3811 | 0.7 |
| No log | 19.0 | 95 | 1.3339 | 0.7 |
| No log | 20.0 | 100 | 1.2919 | 0.75 |
| No log | 21.0 | 105 | 1.2493 | 0.725 |
| No log | 22.0 | 110 | 1.2091 | 0.725 |
| No log | 23.0 | 115 | 1.1707 | 0.75 |
| No log | 24.0 | 120 | 1.1311 | 0.775 |
| No log | 25.0 | 125 | 1.0946 | 0.825 |
| No log | 26.0 | 130 | 1.0642 | 0.8 |
| No log | 27.0 | 135 | 1.0363 | 0.8 |
| No log | 28.0 | 140 | 1.0172 | 0.8 |
| No log | 29.0 | 145 | 0.9939 | 0.825 |
| No log | 30.0 | 150 | 0.9682 | 0.825 |
| No log | 31.0 | 155 | 0.9443 | 0.8 |
| No log | 32.0 | 160 | 0.9289 | 0.825 |
| No log | 33.0 | 165 | 0.9113 | 0.85 |
| No log | 34.0 | 170 | 0.9017 | 0.85 |
| No log | 35.0 | 175 | 0.8804 | 0.85 |
| No log | 36.0 | 180 | 0.8598 | 0.85 |
| No log | 37.0 | 185 | 0.8484 | 0.825 |
| No log | 38.0 | 190 | 0.8361 | 0.825 |
| No log | 39.0 | 195 | 0.8306 | 0.825 |
| No log | 40.0 | 200 | 0.8242 | 0.825 |
| No log | 41.0 | 205 | 0.8198 | 0.85 |
| No log | 42.0 | 210 | 0.8051 | 0.85 |
| No log | 43.0 | 215 | 0.7854 | 0.85 |
| No log | 44.0 | 220 | 0.7727 | 0.9 |
| No log | 45.0 | 225 | 0.7626 | 0.9 |
| No log | 46.0 | 230 | 0.7579 | 0.9 |
| No log | 47.0 | 235 | 0.7500 | 0.9 |
| No log | 48.0 | 240 | 0.7477 | 0.875 |
| No log | 49.0 | 245 | 0.7517 | 0.85 |
| No log | 50.0 | 250 | 0.7484 | 0.85 |
| No log | 51.0 | 255 | 0.7446 | 0.85 |
| No log | 52.0 | 260 | 0.7414 | 0.85 |
| No log | 53.0 | 265 | 0.7357 | 0.875 |
| No log | 54.0 | 270 | 0.7303 | 0.875 |
| No log | 55.0 | 275 | 0.7249 | 0.875 |
| No log | 56.0 | 280 | 0.7232 | 0.875 |
| No log | 57.0 | 285 | 0.7228 | 0.875 |
| No log | 58.0 | 290 | 0.7215 | 0.875 |
| No log | 59.0 | 295 | 0.7206 | 0.875 |
| No log | 60.0 | 300 | 0.7203 | 0.875 |
### Framework versions
- Transformers 4.34.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.14.1
|