Trained on custom dataset

README.md

@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [FacebookAI/roberta-base](https://huggingface.co/FacebookAI/roberta-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3682
-- Precision Macro: 0.8383
-- Recall Macro: 0.8140
-- F1 Macro: 0.8200
+- Loss: 0.3722
+- Precision Macro: 0.8399
+- Recall Macro: 0.8127
+- F1 Macro: 0.8177
 - Accuracy: 0.8265
 
 ## Model description
@@ -47,15 +47,14 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Precision Macro | Recall Macro | F1 Macro | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------------:|:------------:|:--------:|:--------:|
-| 0.4598        | 1.0   | 270  | 0.4176          | 0.8068          | 0.8221       | 0.8125   | 0.8052   |
-| 0.3705        | 2.0   | 540  | 0.3682          | 0.8383          | 0.8140       | 0.8200   | 0.8265   |
-| 0.3           | 3.0   | 810  | 0.4171          | 0.8237          | 0.8122       | 0.8150   | 0.8182   |
+| 0.4766        | 1.0   | 270  | 0.3801          | 0.8110          | 0.8230       | 0.8160   | 0.8089   |
+| 0.3689        | 2.0   | 540  | 0.3722          | 0.8399          | 0.8127       | 0.8177   | 0.8265   |
 
 
 ### Framework versions