update model card README.md

README.md

@@ -22,10 +22,10 @@ model-index:
     metrics:
     - name: F1
       type: f1
-      value: 0.5677001388246182
+      value: 0.5928384279475983
     - name: Accuracy
       type: accuracy
-      value: 0.4480280132694434
+      value: 0.49428676741614447
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -35,10 +35,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on the go_emotions dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0902
-- F1: 0.5677
-- Roc Auc: 0.7357
-- Accuracy: 0.4480
+- Loss: 0.0978
+- F1: 0.5928
+- Roc Auc: 0.7602
+- Accuracy: 0.4943
 
 ## Model description
 
@@ -58,19 +58,21 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 2
+- train_batch_size: 1
 - eval_batch_size: 64
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 4341
-- num_epochs: 1
+- lr_scheduler_warmup_steps: (13023,)
+- num_epochs: 3
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | F1     | Roc Auc | Accuracy |
-|:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|:--------:|
-| 0.0907        | 1.0   | 21705 | 0.0902          | 0.5677 | 0.7357  | 0.4480   |
+| Training Loss | Epoch | Step   | Validation Loss | F1     | Roc Auc | Accuracy |
+|:-------------:|:-----:|:------:|:---------------:|:------:|:-------:|:--------:|
+| 0.1086        | 1.0   | 43410  | 0.0993          | 0.5690 | 0.7444  | 0.4740   |
+| 0.1003        | 2.0   | 86820  | 0.0969          | 0.5885 | 0.7552  | 0.4893   |
+| 0.0858        | 3.0   | 130230 | 0.0978          | 0.5928 | 0.7602  | 0.4943   |
 
 
 ### Framework versions