Upload TFBertForSequenceClassification

README.md

@@ -13,7 +13,11 @@ probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on an unknown dataset.
 It achieves the following results on the evaluation set:
-
+- Train Loss: 0.1618
+- Train Accuracy: 0.9368
+- Validation Loss: 0.2226
+- Validation Accuracy: 0.9102
+- Epoch: 2
 
 ## Model description
 
@@ -32,14 +36,16 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- optimizer: None
+- 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': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 2214, 'end_learning_rate': 1e-05, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
 - training_precision: float32
 
 ### Training results
-It achieved the following accuracy during training (on validation sets):  
-SNLI: 87,90%  
-MNLI: 82.27%  
-ANLI_r3: 44.83%  
+
+| Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
+|:----------:|:--------------:|:---------------:|:-------------------:|:-----:|
+| 0.3017     | 0.8710         | 0.2638          | 0.8843              | 0     |
+| 0.1989     | 0.9198         | 0.2287          | 0.9139              | 1     |
+| 0.1618     | 0.9368         | 0.2226          | 0.9102              | 2     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e567ca24afc2498079d5887e4bbc181128eaea99603b4105f613397d1b776076
+oid sha256:bfe51073c1ae69cb2dba99c9c5ea26218735d273c4875b90eebab6c839550d67
 size 51152184