gabriellabollici commited on
Commit
19e3b45
1 Parent(s): 7316b08

update model card README.md

Browse files
Files changed (1) hide show
  1. mbart-neutralization/README.md +62 -0
mbart-neutralization/README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - simplification
5
+ - generated_from_trainer
6
+ metrics:
7
+ - bleu
8
+ model-index:
9
+ - name: mbart-neutralization
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # mbart-neutralization
17
+
18
+ This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.0105
21
+ - Bleu: 99.063
22
+ - Gen Len: 18.5729
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 5.6e-05
42
+ - train_batch_size: 8
43
+ - eval_batch_size: 8
44
+ - seed: 42
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - num_epochs: 2
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
52
+ |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
53
+ | No log | 1.0 | 440 | 0.0636 | 97.8966 | 19.0833 |
54
+ | 0.2349 | 2.0 | 880 | 0.0105 | 99.063 | 18.5729 |
55
+
56
+
57
+ ### Framework versions
58
+
59
+ - Transformers 4.26.1
60
+ - Pytorch 1.13.1+cu116
61
+ - Datasets 2.10.1
62
+ - Tokenizers 0.13.2