update model card README.md

README.md

@@ -15,13 +15,13 @@ model-index:
     dataset:
       name: opus100
       type: opus100
-      config: en-mr
+      config: en-ta
       split: test
-      args: en-mr
+      args: en-ta
     metrics:
     - name: Bleu
       type: bleu
-      value: 18.5586
+      value: 14.1043
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,9 +31,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [ShyamVarahagiri/MachineTranslation](https://huggingface.co/ShyamVarahagiri/MachineTranslation) on the opus100 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.8066
-- Bleu: 18.5586
-- Gen Len: 9.8455
+- Loss: 2.2734
+- Bleu: 14.1043
+- Gen Len: 11.8405
 
 ## Model description
 
@@ -60,15 +60,14 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 768
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 2
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
-| No log        | 0.99  | 35   | 1.9316          | 15.4969 | 9.799   |
-| No log        | 1.99  | 70   | 1.8314          | 17.8296 | 9.988   |
-| No log        | 2.98  | 105  | 1.8066          | 18.5586 | 9.8455  |
+| No log        | 1.0   | 295  | 2.3233          | 13.2366 | 12.295  |
+| 2.6303        | 2.0   | 590  | 2.2734          | 14.1043 | 11.8405 |
 
 
 ### Framework versions