eldor-97
/

MarianMix_en-10

Text2Text Generation

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

eldor-97 commited on Jan 30, 2022

Commit

0da8030

·

1 Parent(s): 97fafa2

update model card README.md

Files changed (1) hide show

README.md +14 -14

README.md CHANGED Viewed

@@ -16,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [Helsinki-NLP/opus-tatoeba-en-ja](https://huggingface.co/Helsinki-NLP/opus-tatoeba-en-ja) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6226
-- Bleu: 32.5826
-- Gen Len: 35.0804
 ## Model description
@@ -51,17 +51,17 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
-| 1.9496        | 0.44  | 500  | 1.5668          | 1.823   | 41.7538 |
-| 0.9039        | 0.89  | 1000 | 1.2438          | 7.6743  | 40.609  |
-| 0.7176        | 1.33  | 1500 | 1.0046          | 14.2826 | 41.2442 |
-| 0.6161        | 1.78  | 2000 | 0.8634          | 20.3373 | 37.4141 |
-| 0.5239        | 2.22  | 2500 | 0.7740          | 24.5708 | 36.9407 |
-| 0.4689        | 2.67  | 3000 | 0.7182          | 27.3402 | 34.5799 |
-| 0.4286        | 3.11  | 3500 | 0.6859          | 28.7385 | 34.4965 |
-| 0.3852        | 3.56  | 4000 | 0.6598          | 30.1959 | 34.4131 |
-| 0.3845        | 4.0   | 4500 | 0.6397          | 31.6416 | 35.1799 |
-| 0.344         | 4.45  | 5000 | 0.6310          | 32.0065 | 34.8492 |
-| 0.3372        | 4.89  | 5500 | 0.6226          | 32.5826 | 35.0804 |
 ### Framework versions

 This model is a fine-tuned version of [Helsinki-NLP/opus-tatoeba-en-ja](https://huggingface.co/Helsinki-NLP/opus-tatoeba-en-ja) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.8545
+- Bleu: 24.9113
+- Gen Len: 39.1404
 ## Model description
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
+| 2.0876        | 0.44  | 500  | 1.8831          | 0.7013  | 67.3039 |
+| 1.0798        | 0.89  | 1000 | 1.5775          | 3.3504  | 62.6921 |
+| 0.8689        | 1.33  | 1500 | 1.2670          | 10.666  | 38.6369 |
+| 0.7569        | 1.78  | 2000 | 1.1052          | 15.5126 | 39.0221 |
+| 0.6616        | 2.22  | 2500 | 1.0188          | 18.7149 | 39.7563 |
+| 0.5975        | 2.67  | 3000 | 0.9571          | 20.8306 | 38.6138 |
+| 0.5601        | 3.11  | 3500 | 0.9231          | 22.1631 | 38.4594 |
+| 0.5162        | 3.56  | 4000 | 0.8886          | 23.4946 | 38.4875 |
+| 0.5056        | 4.0   | 4500 | 0.8747          | 24.1626 | 38.7101 |
+| 0.4664        | 4.45  | 5000 | 0.8608          | 24.7028 | 38.6409 |
+| 0.4668        | 4.89  | 5500 | 0.8545          | 24.9113 | 39.1404 |
 ### Framework versions