End of training

README.md

@@ -7,6 +7,7 @@ datasets:
 - emotion
 metrics:
 - accuracy
+- f1
 model-index:
 - name: bert-base-uncased-finetuned-emotion
   results:
@@ -23,7 +24,11 @@ model-index:
     - name: Accuracy
       type: accuracy
       value:
-        accuracy: 0.9385
+        accuracy: 0.9355
+    - name: F1
+      type: f1
+      value:
+        f1: 0.935388774713548
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +38,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the emotion dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1772
-- Accuracy: {'accuracy': 0.9385}
+- Loss: 0.1651
+- Accuracy: {'accuracy': 0.9355}
+- F1: {'f1': 0.935388774713548}
 
 ## Model description
 
@@ -63,11 +69,11 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy             |
-|:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.2449        | 1.0   | 1000 | 0.1787          | {'accuracy': 0.9355} |
-| 0.1425        | 2.0   | 2000 | 0.1780          | {'accuracy': 0.9355} |
-| 0.092         | 3.0   | 3000 | 0.1772          | {'accuracy': 0.9385} |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy             | F1                         |
+|:-------------:|:-----:|:----:|:---------------:|:--------------------:|:--------------------------:|
+| 0.2519        | 1.0   | 1000 | 0.1878          | {'accuracy': 0.9325} | {'f1': 0.9323540471733189} |
+| 0.1434        | 2.0   | 2000 | 0.1799          | {'accuracy': 0.9335} | {'f1': 0.9341179573678701} |
+| 0.0907        | 3.0   | 3000 | 0.1651          | {'accuracy': 0.9355} | {'f1': 0.935388774713548}  |
 
 
 ### Framework versions