Model save

README.md

@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Helsinki-NLP/opus-mt-en-fr](https://huggingface.co/Helsinki-NLP/opus-mt-en-fr) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.3017
-- Bleu: 31.7868
-- Gen Len: 29.5263
+- Loss: 1.2652
+- Bleu: 32.1335
+- Gen Len: 29.421
 
 ## Model description
 
@@ -39,8 +39,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 32
+- train_batch_size: 128
+- eval_batch_size: 128
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
@@ -50,13 +50,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|
-| 0.9366        | 1.0   | 31253 | 1.3011          | 31.6097 | 29.5003 |
-| 0.8866        | 2.0   | 62506 | 1.3017          | 31.7868 | 29.5263 |
+| 0.9379        | 1.0   | 7813  | 1.2654          | 32.3231 | 29.4617 |
+| 0.9105        | 2.0   | 15626 | 1.2652          | 32.1335 | 29.421  |
 
 
 ### Framework versions
 
 - Transformers 4.32.1
 - Pytorch 1.12.1
-- Datasets 2.18.0
+- Datasets 2.21.0
 - Tokenizers 0.13.2

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0a7787510fbd2dd8e4cad658f8839206d7169df3d3d074a4fa630c2eb7b1f9f
+oid sha256:ae4364f49b5a53951249aad0b62db42595a6b6ec8eb8f3d31e278fbd727d604a
 size 298761157