Hugging Face's logo

celinely
/
camembert-base-finetuned-sentence-simplification-fr

Text2Text Generation
Transformers
PyTorch
TensorBoard
encoder-decoder
generated_from_trainer
Inference Endpoints
Model card Files Metrics Community
6
celinely commited on
Commit
5664c55
•
1 Parent(s): 4690ed2

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -11
README.md CHANGED
@@ -16,11 +16,11 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.5912
20
- - Rouge1: 56.6081
21
- - Rouge2: 31.7858
22
- - Rougel: 48.4959
23
- - Rougelsum: 56.6806
24
 
25
  ## Model description
26
 
@@ -40,21 +40,28 @@ More information needed
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 2e-05
43
- - train_batch_size: 32
44
- - eval_batch_size: 32
45
  - seed: 42
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
- - num_epochs: 3
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
54
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
55
- | 5.3993 | 1.0 | 188 | 1.4868 | 23.2843 | 6.9787 | 22.46 | 23.2664 |
56
- | 1.2384 | 2.0 | 376 | 0.7450 | 47.5051 | 23.5773 | 42.2383 | 47.5388 |
57
- | 0.8417 | 3.0 | 564 | 0.5912 | 56.6081 | 31.7858 | 48.4959 | 56.6806 |
 
 
 
 
 
 
 
58
 
59
 
60
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.0225
20
+ - Rouge1: 98.9126
21
+ - Rouge2: 96.9479
22
+ - Rougel: 97.9209
23
+ - Rougelsum: 98.9061
24
 
25
  ## Model description
26
 
 
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 2e-05
43
+ - train_batch_size: 16
44
+ - eval_batch_size: 16
45
  - seed: 42
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
+ - num_epochs: 10
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
54
  |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|
55
+ | 3.2555 | 1.0 | 375 | 0.7613 | 41.6446 | 20.4343 | 38.0279 | 41.5954 |
56
+ | 0.679 | 2.0 | 750 | 0.3463 | 72.8071 | 48.9808 | 60.7026 | 72.8052 |
57
+ | 0.4088 | 3.0 | 1125 | 0.1948 | 85.3976 | 65.3267 | 74.3572 | 85.3705 |
58
+ | 0.2795 | 4.0 | 1500 | 0.1098 | 91.8037 | 78.9948 | 85.9716 | 91.8695 |
59
+ | 0.204 | 5.0 | 1875 | 0.0776 | 94.6475 | 84.3954 | 89.9382 | 94.6349 |
60
+ | 0.1544 | 6.0 | 2250 | 0.0454 | 97.197 | 91.932 | 94.8966 | 97.1919 |
61
+ | 0.1212 | 7.0 | 2625 | 0.0384 | 97.5777 | 93.2443 | 95.4839 | 97.5692 |
62
+ | 0.1037 | 8.0 | 3000 | 0.0315 | 97.8918 | 95.2195 | 96.8449 | 97.9063 |
63
+ | 0.0942 | 9.0 | 3375 | 0.0253 | 98.6234 | 96.5271 | 97.6489 | 98.6284 |
64
+ | 0.0823 | 10.0 | 3750 | 0.0225 | 98.9126 | 96.9479 | 97.9209 | 98.9061 |
65
 
66
 
67
  ### Framework versions