Hugging Face's logo

cor-c
/
layoutlm-cord-tf

Token Classification
Transformers
TensorFlow
layoutlm
generated_from_keras_callback
Inference Endpoints
Model card Files Community
cor-c commited on
Commit
7302762
•
1 Parent(s): 2b03620

Upload TFLayoutLMForTokenClassification

Browse files
Files changed (2) hide show
  1. README.md +15 -15
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -1,8 +1,8 @@
1
  ---
2
  license: mit
 
3
  tags:
4
  - generated_from_keras_callback
5
- base_model: microsoft/layoutlm-base-uncased
6
  model-index:
7
  - name: layoutlm-cord-tf
8
  results: []
@@ -15,12 +15,12 @@ probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Train Loss: 0.0418
19
- - Validation Loss: 0.1882
20
- - Train Overall Precision: 0.9468
21
- - Train Overall Recall: 0.9490
22
- - Train Overall F1: 0.9479
23
- - Train Overall Accuracy: 0.9652
24
  - Epoch: 7
25
 
26
  ## Model description
@@ -47,14 +47,14 @@ The following hyperparameters were used during training:
47
 
48
  | Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
49
  |:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
50
- | 1.3129 | 0.5622 | 0.8065 | 0.8120 | 0.8093 | 0.8579 | 0 |
51
- | 0.3927 | 0.2756 | 0.8965 | 0.8965 | 0.8965 | 0.9291 | 1 |
52
- | 0.2136 | 0.1945 | 0.9233 | 0.9254 | 0.9244 | 0.9499 | 2 |
53
- | 0.1335 | 0.1814 | 0.9363 | 0.9399 | 0.9381 | 0.9550 | 3 |
54
- | 0.0915 | 0.1956 | 0.9390 | 0.9482 | 0.9436 | 0.9571 | 4 |
55
- | 0.0810 | 0.2157 | 0.9392 | 0.9399 | 0.9395 | 0.9559 | 5 |
56
- | 0.0628 | 0.1726 | 0.9410 | 0.9467 | 0.9439 | 0.9618 | 6 |
57
- | 0.0418 | 0.1882 | 0.9468 | 0.9490 | 0.9479 | 0.9652 | 7 |
58
 
59
 
60
  ### Framework versions
 
1
  ---
2
  license: mit
3
+ base_model: microsoft/layoutlm-base-uncased
4
  tags:
5
  - generated_from_keras_callback
 
6
  model-index:
7
  - name: layoutlm-cord-tf
8
  results: []
 
15
 
16
  This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.0379
19
+ - Validation Loss: 0.1673
20
+ - Train Overall Precision: 0.9428
21
+ - Train Overall Recall: 0.9528
22
+ - Train Overall F1: 0.9478
23
+ - Train Overall Accuracy: 0.9605
24
  - Epoch: 7
25
 
26
  ## Model description
 
47
 
48
  | Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
49
  |:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
50
+ | 1.2812 | 0.5563 | 0.8209 | 0.8265 | 0.8237 | 0.8574 | 0 |
51
+ | 0.3842 | 0.2602 | 0.9121 | 0.9155 | 0.9138 | 0.9406 | 1 |
52
+ | 0.2075 | 0.2085 | 0.9257 | 0.9201 | 0.9229 | 0.9453 | 2 |
53
+ | 0.1377 | 0.1793 | 0.9368 | 0.9368 | 0.9368 | 0.9555 | 3 |
54
+ | 0.0889 | 0.1902 | 0.9282 | 0.9346 | 0.9314 | 0.9525 | 4 |
55
+ | 0.0747 | 0.1610 | 0.9455 | 0.9513 | 0.9484 | 0.9648 | 5 |
56
+ | 0.0514 | 0.1796 | 0.9480 | 0.9566 | 0.9523 | 0.9639 | 6 |
57
+ | 0.0379 | 0.1673 | 0.9428 | 0.9528 | 0.9478 | 0.9605 | 7 |
58
 
59
 
60
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:08a2c635b84424de0ef5685391e6bb44cbfd39024c16ccfa206ddbc08ff121a5
3
  size 450878408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92f9a0c707e792528248ebdd551a4ac0ca1a4f21e09947f5d8ca0d47591b52ff
3
  size 450878408