Upload TFBertForTokenClassification

README.md

@@ -14,9 +14,9 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.0080
-- Validation Loss: 0.2646
-- Epoch: 12
+- Train Loss: 0.0087
+- Validation Loss: 0.2588
+- Epoch: 9
 
 ## Model description
 
@@ -35,26 +35,23 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 28900, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
+- optimizer: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 2890, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
 - training_precision: mixed_float16
 
 ### Training results
 
 | Train Loss | Validation Loss | Epoch |
 |:----------:|:---------------:|:-----:|
-| 0.4750     | 0.1843          | 0     |
-| 0.1754     | 0.2091          | 1     |
-| 0.1081     | 0.1819          | 2     |
-| 0.0658     | 0.2115          | 3     |
-| 0.0463     | 0.2220          | 4     |
-| 0.0281     | 0.1990          | 5     |
-| 0.0268     | 0.2807          | 6     |
-| 0.0249     | 0.3543          | 7     |
-| 0.0178     | 0.2623          | 8     |
-| 0.0122     | 0.3002          | 9     |
-| 0.0137     | 0.2659          | 10    |
-| 0.0111     | 0.2615          | 11    |
-| 0.0080     | 0.2646          | 12    |
+| 0.5009     | 0.2286          | 0     |
+| 0.1843     | 0.1930          | 1     |
+| 0.1116     | 0.1836          | 2     |
+| 0.0686     | 0.1865          | 3     |
+| 0.0430     | 0.2524          | 4     |
+| 0.0311     | 0.2881          | 5     |
+| 0.0186     | 0.2460          | 6     |
+| 0.0156     | 0.2437          | 7     |
+| 0.0127     | 0.2700          | 8     |
+| 0.0087     | 0.2588          | 9     |
 
 
 ### Framework versions

config.json

@@ -13,7 +13,7 @@
     "0": "O",
     "1": "B-Brand",
     "10": "I-TM",
-    "2": "I-brand",
+    "2": "I-Brand",
     "3": "B-Model",
     "4": "I-Model",
     "5": "B-Tipe",
@@ -30,11 +30,11 @@
     "B-Model": "3",
     "B-TM": "9",
     "B-Tipe": "5",
+    "I-Brand": "2",
     "I-CC": "8",
     "I-Model": "4",
     "I-TM": "10",
     "I-Tipe": "6",
-    "I-brand": "2",
     "O": "0"
   },
   "layer_norm_eps": 1e-12,

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c4bb67c4c70e979989717ab5ab5df35b00b3eb7cfad26f3f4b8930e72e4fe83
+oid sha256:d7dd0afd67966a8f4adf4668b57b2e0e8307a7843264002c1965f82b94115b55
 size 431198260