init

README.md

@@ -14,11 +14,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 # opus-mt-en-st
 
-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.
+This model is a fine-tuned version of [cw1521/opus-mt-en-st](https://huggingface.co/cw1521/opus-mt-en-st) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3835
-- Bleu: 77.0555
-- Gen Len: 60.4371
+- Loss: 0.3710
+- Bleu: 77.1725
+- Gen Len: 60.3696
 
 ## Model description
 
@@ -49,7 +49,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Bleu    | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|
-| 0.5985        | 1.0   | 969  | 0.3835          | 77.0555 | 60.4371 |
+| 0.3768        | 1.0   | 969  | 0.3710          | 77.1725 | 60.3696 |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "Helsinki-NLP/opus-mt-en-fr",
+  "_name_or_path": "cw1521/opus-mt-en-st",
   "_num_labels": 3,
   "activation_dropout": 0.0,
   "activation_function": "swish",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2922369a2976cd4ca818fc44ece4589609135a549326d87029c8caeae4cdac9f
+oid sha256:ed7e1a32539f6f7f1ecb372537af824e93f48d58c5006fd8bc28d6202bb23d0c
 size 298757573

tokenizer_config.json

@@ -1,7 +1,7 @@
 {
   "eos_token": "</s>",
   "model_max_length": 512,
-  "name_or_path": "Helsinki-NLP/opus-mt-en-fr",
+  "name_or_path": "cw1521/opus-mt-en-st",
   "pad_token": "<pad>",
   "separate_vocabs": false,
   "source_lang": "en",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99c324cbfd7ff38feef11d103c8233b91085623eddd6f62b58e1f4cd2e481136
+oid sha256:3c3de9648df87ac76e89f11ae40328b988c77c44b33f276a23dad50979e56066
 size 3503