End of training

README.md

@@ -17,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the wmt20_mlqe_task1 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5558
-- R Squared: 0.0026
-- Mae: 0.5879
-- Pearson R: 0.1407
+- Loss: 0.5090
+- R Squared: 0.0865
+- Mae: 0.5291
+- Pearson R: 0.3627
 
 ## Model description
 
@@ -42,18 +42,20 @@ The following hyperparameters were used during training:
 - learning_rate: 2e-05
 - train_batch_size: 16
 - eval_batch_size: 16
-- seed: 1985
+- seed: 1986
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | R Squared | Mae    | Pearson R |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:---------:|
-| No log        | 1.0   | 438  | 0.5630          | -0.0103   | 0.5932 | 0.1633    |
-| 0.703         | 2.0   | 876  | 0.5609          | -0.0066   | 0.5935 | 0.1863    |
-| 0.7258        | 3.0   | 1314 | 0.5558          | 0.0026    | 0.5879 | 0.1407    |
+| No log        | 1.0   | 438  | 0.5544          | 0.0051    | 0.5990 | 0.3021    |
+| 0.6821        | 2.0   | 876  | 0.5527          | 0.0082    | 0.5998 | 0.1601    |
+| 0.7102        | 3.0   | 1314 | 0.5400          | 0.0309    | 0.5712 | 0.3027    |
+| 0.7194        | 4.0   | 1752 | 0.5132          | 0.0791    | 0.5401 | 0.3557    |
+| 0.6285        | 5.0   | 2190 | 0.5090          | 0.0865    | 0.5291 | 0.3627    |
 
 
 ### Framework versions