Commit
•
6c3286d
1
Parent(s):
274603b
Upload TFLayoutLMForTokenClassification
Browse files- README.md +17 -10
- 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-funsd-tf
|
8 |
results: []
|
@@ -15,13 +15,13 @@ 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:
|
19 |
-
- Validation Loss:
|
20 |
-
- Train Overall Precision: 0.
|
21 |
-
- Train Overall Recall: 0.
|
22 |
-
- Train Overall F1: 0.
|
23 |
-
- Train Overall Accuracy: 0.
|
24 |
-
- Epoch:
|
25 |
|
26 |
## Model description
|
27 |
|
@@ -40,14 +40,21 @@ More information needed
|
|
40 |
### Training hyperparameters
|
41 |
|
42 |
The following hyperparameters were used during training:
|
43 |
-
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate':
|
44 |
- training_precision: mixed_float16
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
|
49 |
|:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
|
50 |
-
| 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
|
52 |
|
53 |
### 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-funsd-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: 0.2406
|
19 |
+
- Validation Loss: 0.7155
|
20 |
+
- Train Overall Precision: 0.7459
|
21 |
+
- Train Overall Recall: 0.7893
|
22 |
+
- Train Overall F1: 0.7669
|
23 |
+
- Train Overall Accuracy: 0.8064
|
24 |
+
- Epoch: 7
|
25 |
|
26 |
## Model description
|
27 |
|
|
|
40 |
### Training hyperparameters
|
41 |
|
42 |
The following hyperparameters were used during training:
|
43 |
+
- optimizer: {'inner_optimizer': {'module': 'transformers.optimization_tf', 'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': 2.9999999242136255e-05, 'decay': 0.0, 'beta_1': 0.8999999761581421, 'beta_2': 0.9990000128746033, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}, 'registered_name': 'AdamWeightDecay'}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
|
44 |
- training_precision: mixed_float16
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Train Loss | Validation Loss | Train Overall Precision | Train Overall Recall | Train Overall F1 | Train Overall Accuracy | Epoch |
|
49 |
|:----------:|:---------------:|:-----------------------:|:--------------------:|:----------------:|:----------------------:|:-----:|
|
50 |
+
| 1.7115 | 1.4279 | 0.2575 | 0.2965 | 0.2757 | 0.4884 | 0 |
|
51 |
+
| 1.1520 | 0.8490 | 0.5994 | 0.6854 | 0.6395 | 0.7372 | 1 |
|
52 |
+
| 0.7816 | 0.7069 | 0.6391 | 0.7471 | 0.6889 | 0.7808 | 2 |
|
53 |
+
| 0.5815 | 0.6601 | 0.7089 | 0.7672 | 0.7369 | 0.7992 | 3 |
|
54 |
+
| 0.4460 | 0.6306 | 0.7093 | 0.7787 | 0.7424 | 0.8060 | 4 |
|
55 |
+
| 0.3658 | 0.6575 | 0.7372 | 0.7812 | 0.7586 | 0.8111 | 5 |
|
56 |
+
| 0.2926 | 0.6658 | 0.7240 | 0.7832 | 0.7525 | 0.8096 | 6 |
|
57 |
+
| 0.2406 | 0.7155 | 0.7459 | 0.7893 | 0.7669 | 0.8064 | 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:
|
3 |
size 450829256
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:256e2c8dca24bfe0d096efbbf919b28b972daa6a01c8026d297ff72ecc30b58f
|
3 |
size 450829256
|