Hugging Face's logo

gagan3012
/
ArOCR

Image-to-Text
Transformers
PyTorch
TensorBoard
Arabic
vision-encoder-decoder
image-text-to-text
Eval Results (legacy)
Model card Files
xet
 Metrics Community
3
gagan3012 commited on
Commit
8a8ea42
·
1 Parent(s): 0bf8c4e

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -27
README.md CHANGED
@@ -1,18 +1,9 @@
1
  ---
2
  tags:
3
  - generated_from_trainer
4
- - trocr
5
- language: ar
6
  model-index:
7
  - name: ArOCR
8
- results:
9
- - task:
10
- name: Optical Charater Recogntion
11
- type: image-to-text
12
- metrics:
13
- - name: Test CER
14
- type: cer
15
- value: 4.46
16
  ---
17
 
18
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -22,8 +13,8 @@ should probably proofread and complete it, then remove this comment. -->
22
 
23
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
24
  It achieves the following results on the evaluation set:
25
- - Loss: 0.0620
26
- - Cer: 0.0446
27
 
28
  ## Model description
29
 
@@ -43,28 +34,26 @@ More information needed
43
 
44
  The following hyperparameters were used during training:
45
  - learning_rate: 5e-05
46
- - train_batch_size: 4
47
- - eval_batch_size: 4
48
  - seed: 42
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: linear
51
- - num_epochs: 3.0
52
  - mixed_precision_training: Native AMP
53
 
54
  ### Training results
55
 
56
- | Training Loss | Epoch | Step | Validation Loss | Cer |
57
- |:-------------:|:-----:|:-----:|:---------------:|:------:|
58
- | 3.6633 | 0.3 | 1000 | 3.3381 | 1.0385 |
59
- | 2.3915 | 0.59 | 2000 | 2.3581 | 0.6389 |
60
- | 1.5061 | 0.89 | 3000 | 1.4635 | 0.4779 |
61
- | 0.8043 | 1.18 | 4000 | 0.8260 | 0.3427 |
62
- | 0.46 | 1.48 | 5000 | 0.5074 | 0.2242 |
63
- | 0.3393 | 1.77 | 6000 | 0.2699 | 0.1219 |
64
- | 0.1077 | 2.07 | 7000 | 0.1794 | 0.0933 |
65
- | 0.063 | 2.37 | 8000 | 0.1343 | 0.0617 |
66
- | 0.0356 | 2.66 | 9000 | 0.0790 | 0.0692 |
67
- | 0.0292 | 2.96 | 10000 | 0.0620 | 0.0446 |
68
 
69
 
70
  ### Framework versions
 
1
  ---
2
  tags:
3
  - generated_from_trainer
 
 
4
  model-index:
5
  - name: ArOCR
6
+ results: []
 
 
 
 
 
 
 
7
  ---
8
 
9
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 0.0407
17
+ - Cer: 0.0200
18
 
19
  ## Model description
20
 
 
34
 
35
  The following hyperparameters were used during training:
36
  - learning_rate: 5e-05
37
+ - train_batch_size: 8
38
+ - eval_batch_size: 8
39
  - seed: 42
40
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
41
  - lr_scheduler_type: linear
42
+ - num_epochs: 5
43
  - mixed_precision_training: Native AMP
44
 
45
  ### Training results
46
 
47
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
48
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
49
+ | 1.6164 | 0.59 | 1000 | 1.4109 | 0.5793 |
50
+ | 0.3434 | 1.18 | 2000 | 0.3876 | 0.2176 |
51
+ | 0.1679 | 1.77 | 3000 | 0.2262 | 0.1186 |
52
+ | 0.0816 | 2.37 | 4000 | 0.1274 | 0.0634 |
53
+ | 0.0421 | 2.96 | 5000 | 0.0817 | 0.0381 |
54
+ | 0.0067 | 3.55 | 6000 | 0.0520 | 0.0265 |
55
+ | 0.0044 | 4.14 | 7000 | 0.0469 | 0.0215 |
56
+ | 0.0027 | 4.73 | 8000 | 0.0407 | 0.0200 |
 
 
57
 
58
 
59
  ### Framework versions