updates
Browse files
README.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
|
| 3 |
The original instructions are [here](https://github.com/rsennrich/wmt16-scripts/tree/master/sample).
|
| 4 |
|
| 5 |
-
|
| 6 |
|
| 7 |
```
|
| 8 |
git clone https://github.com/rsennrich/wmt16-scripts
|
|
@@ -11,10 +11,9 @@ cd sample
|
|
| 11 |
./download_files.sh
|
| 12 |
./preprocess.sh
|
| 13 |
```
|
| 14 |
-
https://github.com/rsennrich/wmt16-scripts/tree/master/sample
|
| 15 |
|
| 16 |
It was originally used by `transformers` [`finetune_trainer.py`](https://github.com/huggingface/transformers/blob/641f418e102218c4bf16fcd3124bfebed6217ef6/examples/seq2seq/finetune_trainer.py)
|
| 17 |
|
| 18 |
-
The data itself resides https://cdn-datasets.huggingface.co/translation/wmt_en_ro.tar.gz
|
| 19 |
|
| 20 |
If you would like to convert it to jsonlines I've included a small script `convert-to-jsonlines.py` that will do it for you. But if you're using the `datasets` API, it will be done on the fly.
|
|
|
|
| 2 |
|
| 3 |
The original instructions are [here](https://github.com/rsennrich/wmt16-scripts/tree/master/sample).
|
| 4 |
|
| 5 |
+
This pre-processed dataset was created by running:
|
| 6 |
|
| 7 |
```
|
| 8 |
git clone https://github.com/rsennrich/wmt16-scripts
|
|
|
|
| 11 |
./download_files.sh
|
| 12 |
./preprocess.sh
|
| 13 |
```
|
|
|
|
| 14 |
|
| 15 |
It was originally used by `transformers` [`finetune_trainer.py`](https://github.com/huggingface/transformers/blob/641f418e102218c4bf16fcd3124bfebed6217ef6/examples/seq2seq/finetune_trainer.py)
|
| 16 |
|
| 17 |
+
The data itself resides at https://cdn-datasets.huggingface.co/translation/wmt_en_ro.tar.gz
|
| 18 |
|
| 19 |
If you would like to convert it to jsonlines I've included a small script `convert-to-jsonlines.py` that will do it for you. But if you're using the `datasets` API, it will be done on the fly.
|