update model card README.md

README.md

@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [cardiffnlp/twitter-roberta-base-sentiment](https://huggingface.co/cardiffnlp/twitter-roberta-base-sentiment) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1941
-- Accuracy: 0.6047
+- Loss: 0.8601
+- Accuracy: 0.6117
 
 ## Model description
 
@@ -41,27 +41,19 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 99   | 0.8847          | 0.6028   |
-| No log        | 2.0   | 198  | 0.8396          | 0.6180   |
-| No log        | 3.0   | 297  | 0.8832          | 0.5952   |
-| No log        | 4.0   | 396  | 0.8900          | 0.6174   |
-| No log        | 5.0   | 495  | 0.9782          | 0.625    |
-| No log        | 6.0   | 594  | 1.0478          | 0.6193   |
-| No log        | 7.0   | 693  | 1.0683          | 0.6015   |
-| No log        | 8.0   | 792  | 1.1001          | 0.5984   |
-| No log        | 9.0   | 891  | 1.1578          | 0.6155   |
-| No log        | 10.0  | 990  | 1.1941          | 0.6047   |
+| No log        | 1.0   | 84   | 0.8663          | 0.5914   |
+| No log        | 2.0   | 168  | 0.8601          | 0.6117   |
 
 
 ### Framework versions
 
-- Transformers 4.16.0
+- Transformers 4.16.1
 - Pytorch 1.10.0+cu111
-- Datasets 1.18.1
+- Datasets 1.18.2
 - Tokenizers 0.11.0