Hugging Face's logo

Karthikeya55
/
layoutlm-funsd-sequence-tf

Token Classification
Transformers
TensorFlow
TensorBoard
layoutlm
generated_from_keras_callback
Model card Files
xet
 Metrics Community
Karthikeya55 commited on
Commit
d5a11e9
·
1 Parent(s): 5a51c48

End of training

Browse files
Files changed (3) hide show
  1. README.md +16 -16
  2. config.json +1 -1
  3. tf_model.h5 +1 -1
README.md CHANGED
@@ -11,14 +11,14 @@ probably proofread and complete it, then remove this comment. -->
11
 
12
  # layoutlm-funsd-sequence-tf
13
 
14
- This model is a fine-tuned version of [karthikeya55/layout_lm_fine_tune_funsd_dataset](https://huggingface.co/karthikeya55/layout_lm_fine_tune_funsd_dataset) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
- - Train Loss: 0.0611
17
- - Validation Loss: 0.8456
18
- - Train Overall Precision: 0.7485
19
- - Train Overall Recall: 0.8033
20
- - Train Overall F1: 0.7749
21
- - Train Overall Accuracy: 0.8185
22
  - Epoch: 7
23
 
24
  ## Model description
@@ -38,21 +38,21 @@ More information needed
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
- - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
42
  - training_precision: mixed_float16
43
 
44
  ### Training results
45
 
46
  | Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
47
  |:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
48
- | 0.2305 | 0.7086 | 0.7410 | 0.7968 | 0.7679 | 0.8113 | 0 |
49
- | 0.1870 | 0.6923 | 0.7435 | 0.8073 | 0.7741 | 0.8140 | 1 |
50
- | 0.1411 | 0.7292 | 0.7487 | 0.8088 | 0.7776 | 0.8134 | 2 |
51
- | 0.1163 | 0.7648 | 0.7588 | 0.8144 | 0.7856 | 0.8119 | 3 |
52
- | 0.1033 | 0.8033 | 0.7460 | 0.8048 | 0.7743 | 0.8119 | 4 |
53
- | 0.0805 | 0.8018 | 0.7277 | 0.8058 | 0.7648 | 0.8082 | 5 |
54
- | 0.0718 | 0.8344 | 0.7521 | 0.8128 | 0.7813 | 0.8203 | 6 |
55
- | 0.0611 | 0.8456 | 0.7485 | 0.8033 | 0.7749 | 0.8185 | 7 |
56
 
57
 
58
  ### Framework versions
 
11
 
12
  # layoutlm-funsd-sequence-tf
13
 
14
+ This model is a fine-tuned version of [microsoft/layoutlm-base-uncased](https://huggingface.co/microsoft/layoutlm-base-uncased) on an unknown dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Train Loss: 0.2348
17
+ - Validation Loss: 0.6737
18
+ - Train Overall Precision: 0.7356
19
+ - Train Overall Recall: 0.7998
20
+ - Train Overall F1: 0.7663
21
+ - Train Overall Accuracy: 0.8220
22
  - Epoch: 7
23
 
24
  ## Model description
 
38
  ### Training hyperparameters
39
 
40
  The following hyperparameters were used during training:
41
+ - optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': 3e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
42
  - training_precision: mixed_float16
43
 
44
  ### Training results
45
 
46
  | Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
47
  |:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
48
+ | 1.7150 | 1.4139 | 0.2373 | 0.2860 | 0.2594 | 0.4954 | 0 |
49
+ | 1.1803 | 0.9205 | 0.5676 | 0.6322 | 0.5981 | 0.7008 | 1 |
50
+ | 0.7884 | 0.7100 | 0.6202 | 0.7250 | 0.6685 | 0.7735 | 2 |
51
+ | 0.5877 | 0.6476 | 0.6689 | 0.7662 | 0.7142 | 0.7942 | 3 |
52
+ | 0.4490 | 0.6179 | 0.7133 | 0.8078 | 0.7576 | 0.8066 | 4 |
53
+ | 0.3746 | 0.6305 | 0.7176 | 0.7878 | 0.7510 | 0.8129 | 5 |
54
+ | 0.3082 | 0.6924 | 0.7163 | 0.8018 | 0.7566 | 0.7937 | 6 |
55
+ | 0.2348 | 0.6737 | 0.7356 | 0.7998 | 0.7663 | 0.8220 | 7 |
56
 
57
 
58
  ### Framework versions
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "karthikeya55/layout_lm_fine_tune_funsd_dataset",
3
  "architectures": [
4
  "LayoutLMForTokenClassification"
5
  ],
 
1
  {
2
+ "_name_or_path": "microsoft/layoutlm-base-uncased",
3
  "architectures": [
4
  "LayoutLMForTokenClassification"
5
  ],
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbe85a98cda75399e901b982b0347e4079bf9d9b326734695f4a19723d518c5f
3
  size 450829256
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e14477d5f65e179e4682116bf5eb7b6ce423a36c4ce81ad05a504fdd146f844e
3
  size 450829256