update model card README.md

README.md

@@ -3,8 +3,8 @@ license: mit
 tags:
 - generated_from_trainer
 metrics:
-- f1
 - accuracy
+- f1
 model-index:
 - name: roberta-base-finetuned-jigsaw-toxic
   results: []
@@ -17,10 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0412
-- F1: 0.7908
-- Roc Auc: 0.9048
-- Accuracy: 0.9257
+- Loss: 0.0859
+- Accuracy: 0.9747
+- F1: 0.9746
 
 ## Model description
 
@@ -40,24 +39,25 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 48
-- eval_batch_size: 48
+- train_batch_size: 128
+- eval_batch_size: 128
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 2
+- num_epochs: 3
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | F1     | Roc Auc | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
-| 0.0524        | 1.0   | 2774 | 0.0432          | 0.7805 | 0.8940  | 0.9254   |
-| 0.0348        | 2.0   | 5548 | 0.0412          | 0.7908 | 0.9048  | 0.9257   |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
+| 0.1179        | 1.0   | 2116 | 0.0982          | 0.9694   | 0.9694 |
+| 0.0748        | 2.0   | 4232 | 0.0859          | 0.9747   | 0.9746 |
+| 0.0582        | 3.0   | 6348 | 0.0916          | 0.9750   | 0.9750 |
 
 
 ### Framework versions
 
-- Transformers 4.21.0
-- Pytorch 1.12.0+cu113
-- Datasets 2.4.0
+- Transformers 4.22.1
+- Pytorch 1.12.1+cu113
+- Datasets 2.5.1
 - Tokenizers 0.12.1