Model save

README.md

@@ -20,8 +20,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on the emotion dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0773
-- Accuracy: 0.594
+- Loss: 0.1321
+- Accuracy: 0.937
 
 ## Model description
 
@@ -40,20 +40,23 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
+- learning_rate: 0.0005
 - train_batch_size: 32
 - eval_batch_size: 32
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 2
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 2.3952        | 1.0   | 500  | 1.2924          | 0.537    |
-| 1.2039        | 2.0   | 1000 | 1.0773          | 0.594    |
+| 0.6485        | 1.0   | 500  | 0.2474          | 0.914    |
+| 0.2247        | 2.0   | 1000 | 0.1730          | 0.9285   |
+| 0.1752        | 3.0   | 1500 | 0.1544          | 0.9265   |
+| 0.1479        | 4.0   | 2000 | 0.1401          | 0.9345   |
+| 0.1336        | 5.0   | 2500 | 0.1321          | 0.937    |
 
 
 ### Framework versions