Hugging Face's logo

ahmeddbahaa
/
AraBART-finetuned-ar

Summarization
Transformers
PyTorch
TensorBoard
mbart
text2text-generation
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
2
ahmeddbahaa commited on
Commit
c2be33d
•
1 Parent(s): c8d53ae

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -40
README.md CHANGED
@@ -5,22 +5,9 @@ tags:
5
  - generated_from_trainer
6
  datasets:
7
  - xlsum
8
- metrics:
9
- - rouge
10
  model-index:
11
  - name: AraBART-finetuned-ar
12
- results:
13
- - task:
14
- name: Sequence-to-sequence Language Modeling
15
- type: text2text-generation
16
- dataset:
17
- name: xlsum
18
- type: xlsum
19
- args: arabic
20
- metrics:
21
- - name: Rouge1
22
- type: rouge
23
- value: 2.2459
24
  ---
25
 
26
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,12 +17,12 @@ should probably proofread and complete it, then remove this comment. -->
30
 
31
  This model is a fine-tuned version of [moussaKam/AraBART](https://huggingface.co/moussaKam/AraBART) on the xlsum dataset.
32
  It achieves the following results on the evaluation set:
33
- - Loss: 2.3785
34
- - Rouge1: 2.2459
35
- - Rouge2: 0.0
36
- - Rougel: 2.2459
37
- - Rougelsum: 2.2459
38
- - Gen Len: 19.695
39
 
40
  ## Model description
41
 
@@ -55,35 +42,34 @@ More information needed
55
 
56
  The following hyperparameters were used during training:
57
  - learning_rate: 5e-05
58
- - train_batch_size: 4
59
- - eval_batch_size: 4
60
  - seed: 42
61
- - gradient_accumulation_steps: 16
62
- - total_train_batch_size: 64
63
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
64
  - lr_scheduler_type: linear
65
- - lr_scheduler_warmup_ratio: 0.6
66
  - num_epochs: 10
 
67
 
68
  ### Training results
69
 
70
- | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
71
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
72
- | No log | 0.98 | 32 | 2.7774 | 1.0638 | 0.0 | 1.0638 | 1.182 | 19.5177 |
73
- | No log | 1.98 | 64 | 2.4730 | 1.182 | 0.0 | 1.3002 | 1.182 | 19.8121 |
74
- | No log | 2.98 | 96 | 2.4129 | 2.3641 | 0.3546 | 2.3641 | 2.3641 | 19.8298 |
75
- | No log | 3.98 | 128 | 2.3724 | 2.1277 | 0.3546 | 2.1277 | 2.1277 | 19.8121 |
76
- | No log | 4.98 | 160 | 2.3560 | 1.8913 | 0.3546 | 1.8913 | 1.8913 | 19.805 |
77
- | No log | 5.98 | 192 | 2.3574 | 1.5366 | 0.0 | 1.5366 | 1.6548 | 19.7979 |
78
- | No log | 6.98 | 224 | 2.3676 | 2.1277 | 0.3546 | 2.2459 | 2.1277 | 19.6348 |
79
- | No log | 7.98 | 256 | 2.3656 | 2.0095 | 0.0 | 2.0095 | 2.0095 | 19.844 |
80
- | No log | 8.98 | 288 | 2.3751 | 2.2459 | 0.0 | 2.3641 | 2.2459 | 19.6738 |
81
- | No log | 9.98 | 320 | 2.3785 | 2.2459 | 0.0 | 2.2459 | 2.2459 | 19.695 |
82
 
83
 
84
  ### Framework versions
85
 
86
- - Transformers 4.17.0
87
  - Pytorch 1.10.0+cu111
88
- - Datasets 2.0.0
89
- - Tokenizers 0.11.6
 
5
  - generated_from_trainer
6
  datasets:
7
  - xlsum
 
 
8
  model-index:
9
  - name: AraBART-finetuned-ar
10
+ results: []
 
 
 
 
 
 
 
 
 
 
 
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
17
 
18
  This model is a fine-tuned version of [moussaKam/AraBART](https://huggingface.co/moussaKam/AraBART) on the xlsum dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 3.7449
21
+ - Rouge-1: 31.08
22
+ - Rouge-2: 14.68
23
+ - Rouge-l: 27.36
24
+ - Gen Len: 19.64
25
+ - Bertscore: 73.86
26
 
27
  ## Model description
28
 
 
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 5e-05
45
+ - train_batch_size: 16
46
+ - eval_batch_size: 16
47
  - seed: 42
 
 
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
+ - lr_scheduler_warmup_steps: 250
51
  - num_epochs: 10
52
+ - label_smoothing_factor: 0.1
53
 
54
  ### Training results
55
 
56
+ | Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Gen Len | Bertscore |
57
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:---------:|
58
+ | 4.4318 | 1.0 | 2345 | 3.7996 | 28.93 | 13.2 | 25.56 | 19.51 | 73.17 |
59
+ | 4.0338 | 2.0 | 4690 | 3.7483 | 30.29 | 14.24 | 26.73 | 19.5 | 73.59 |
60
+ | 3.8586 | 3.0 | 7035 | 3.7281 | 30.44 | 14.44 | 26.92 | 19.75 | 73.58 |
61
+ | 3.7289 | 4.0 | 9380 | 3.7204 | 30.55 | 14.49 | 26.88 | 19.66 | 73.73 |
62
+ | 3.6245 | 5.0 | 11725 | 3.7199 | 30.73 | 14.63 | 27.11 | 19.69 | 73.68 |
63
+ | 3.5392 | 6.0 | 14070 | 3.7221 | 30.85 | 14.65 | 27.21 | 19.7 | 73.77 |
64
+ | 3.4694 | 7.0 | 16415 | 3.7286 | 31.08 | 14.8 | 27.41 | 19.62 | 73.84 |
65
+ | 3.4126 | 8.0 | 18760 | 3.7384 | 31.06 | 14.77 | 27.41 | 19.64 | 73.82 |
66
+ | 3.3718 | 9.0 | 21105 | 3.7398 | 31.18 | 14.89 | 27.49 | 19.67 | 73.87 |
67
+ | 3.3428 | 10.0 | 23450 | 3.7449 | 31.19 | 14.88 | 27.44 | 19.68 | 73.87 |
68
 
69
 
70
  ### Framework versions
71
 
72
+ - Transformers 4.18.0
73
  - Pytorch 1.10.0+cu111
74
+ - Datasets 2.1.0
75
+ - Tokenizers 0.12.1