Asia-N
/

opus-mt-ar-en-finetuned-ar-to-en

Text2Text Generation

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

Asia-N commited on Jun 17, 2022

Commit

1a5f56a

•

1 Parent(s): ca64fe2

update model card README.md

Files changed (1) hide show

README.md +14 -14

README.md CHANGED Viewed

@@ -19,7 +19,7 @@ model-index:
     metrics:
     - name: Bleu
       type: bleu
-      value: 32.7441
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -29,9 +29,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [Helsinki-NLP/opus-mt-ar-en](https://huggingface.co/Helsinki-NLP/opus-mt-ar-en) on the news_commentary dataset.
 It achieves the following results on the evaluation set:
-- Loss: 8.4400
-- Bleu: 32.7441
-- Gen Len: 57.02
 ## Model description
@@ -62,16 +62,16 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
-| No log        | 1.0   | 4    | 8.4406          | 32.7441 | 57.02   |
-| No log        | 2.0   | 8    | 8.4405          | 32.7441 | 57.02   |
-| No log        | 3.0   | 12   | 8.4403          | 32.7441 | 57.02   |
-| No log        | 4.0   | 16   | 8.4402          | 32.7441 | 57.02   |
-| No log        | 5.0   | 20   | 8.4401          | 32.7441 | 57.02   |
-| No log        | 6.0   | 24   | 8.4400          | 32.7441 | 57.02   |
-| No log        | 7.0   | 28   | 8.4400          | 32.7441 | 57.02   |
-| No log        | 8.0   | 32   | 8.4400          | 32.7441 | 57.02   |
-| No log        | 9.0   | 36   | 8.4400          | 32.7441 | 57.02   |
-| No log        | 10.0  | 40   | 8.4400          | 32.7441 | 57.02   |
 ### Framework versions

     metrics:
     - name: Bleu
       type: bleu
+      value: 32.5319
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 This model is a fine-tuned version of [Helsinki-NLP/opus-mt-ar-en](https://huggingface.co/Helsinki-NLP/opus-mt-ar-en) on the news_commentary dataset.
 It achieves the following results on the evaluation set:
+- Loss: 10.6041
+- Bleu: 32.5319
+- Gen Len: 56.234
 ## Model description
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
+| No log        | 1.0   | 32   | 10.6106         | 32.5327 | 56.234  |
+| No log        | 2.0   | 64   | 10.6090         | 32.5319 | 56.234  |
+| No log        | 3.0   | 96   | 10.6076         | 32.5319 | 56.234  |
+| No log        | 4.0   | 128  | 10.6064         | 32.5319 | 56.234  |
+| No log        | 5.0   | 160  | 10.6053         | 32.5319 | 56.234  |
+| No log        | 6.0   | 192  | 10.6048         | 32.5319 | 56.234  |
+| No log        | 7.0   | 224  | 10.6044         | 32.5319 | 56.234  |
+| No log        | 8.0   | 256  | 10.6042         | 32.5319 | 56.234  |
+| No log        | 9.0   | 288  | 10.6041         | 32.5319 | 56.234  |
+| No log        | 10.0  | 320  | 10.6041         | 32.5319 | 56.234  |
 ### Framework versions