Fine-tuning Complete

README.md

@@ -26,7 +26,7 @@ model-index:
       value: 0.9255
     - name: F1
       type: f1
-      value: 0.9254483411585025
+      value: 0.9254709649783771
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -36,9 +36,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the emotion dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2085
+- Loss: 0.2128
 - Accuracy: 0.9255
-- F1: 0.9254
+- F1: 0.9255
 
 ## Model description
 
@@ -58,22 +58,19 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 256
-- eval_batch_size: 256
+- train_batch_size: 64
+- eval_batch_size: 64
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 5
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
-| 1.3143        | 1.0   | 63   | 0.8964          | 0.711    | 0.6492 |
-| 0.623         | 2.0   | 126  | 0.3847          | 0.889    | 0.8816 |
-| 0.3146        | 3.0   | 189  | 0.2565          | 0.9195   | 0.9189 |
-| 0.2237        | 4.0   | 252  | 0.2168          | 0.926    | 0.9261 |
-| 0.1941        | 5.0   | 315  | 0.2085          | 0.9255   | 0.9254 |
+| 0.5469        | 1.0   | 250  | 0.3084          | 0.906    | 0.9056 |
+| 0.2211        | 2.0   | 500  | 0.2128          | 0.9255   | 0.9255 |
 
 
 ### Framework versions