felixshier commited on
Commit
f563c2a
1 Parent(s): ccce3af

Upload TFBertForSequenceClassification

Browse files
Files changed (2) hide show
  1. README.md +10 -9
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -15,10 +15,10 @@ probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Train Loss: 0.1180
19
- - Validation Loss: 0.4649
20
- - Train F1: 0.8437
21
- - Epoch: 3
22
 
23
  ## Model description
24
 
@@ -37,17 +37,18 @@ More information needed
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
- - optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 4830, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
41
  - training_precision: float32
42
 
43
  ### Training results
44
 
45
  | Train Loss | Validation Loss | Train F1 | Epoch |
46
  |:----------:|:---------------:|:--------:|:-----:|
47
- | 0.4822 | 0.3715 | 0.8490 | 0 |
48
- | 0.3177 | 0.4062 | 0.8283 | 1 |
49
- | 0.2080 | 0.4227 | 0.8441 | 2 |
50
- | 0.1180 | 0.4649 | 0.8437 | 3 |
 
51
 
52
 
53
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.0839
19
+ - Validation Loss: 0.7750
20
+ - Train F1: 0.7755
21
+ - Epoch: 4
22
 
23
  ## Model description
24
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
40
+ - optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 1990, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
41
  - training_precision: float32
42
 
43
  ### Training results
44
 
45
  | Train Loss | Validation Loss | Train F1 | Epoch |
46
  |:----------:|:---------------:|:--------:|:-----:|
47
+ | 0.5526 | 0.6443 | 0.7416 | 0 |
48
+ | 0.3883 | 0.5055 | 0.7648 | 1 |
49
+ | 0.2366 | 0.5839 | 0.7772 | 2 |
50
+ | 0.1339 | 0.6715 | 0.7726 | 3 |
51
+ | 0.0839 | 0.7750 | 0.7755 | 4 |
52
 
53
 
54
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dd2599fa5e6aace10daa8d1d05022197a5ff65f7d01214e9cf0a93363e53f8fd
3
  size 438223128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:767f0b35ee36f1df3f908b534aaa228cd259bd44e64e3772b87fc7fd394abdf6
3
  size 438223128