Update urls
Browse files- completeformer-masked.py +4 -4
completeformer-masked.py
CHANGED
|
@@ -50,11 +50,11 @@ _LICENSE = ""
|
|
| 50 |
# The HuggingFace dataset library don't host the datasets but only point to the original files
|
| 51 |
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
|
| 52 |
_DATA_URLs = {
|
| 53 |
-
"tokenizer": "https://
|
| 54 |
"all": {
|
| 55 |
-
"train": "https://
|
| 56 |
-
"valid": "https://
|
| 57 |
-
"test": "https://
|
| 58 |
},
|
| 59 |
"mix": {
|
| 60 |
"train": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/training_mix.csv",
|
| 50 |
# The HuggingFace dataset library don't host the datasets but only point to the original files
|
| 51 |
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
|
| 52 |
_DATA_URLs = {
|
| 53 |
+
"tokenizer": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/data_for_tokenizer_training.csv",
|
| 54 |
"all": {
|
| 55 |
+
"train": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/training_clean.csv",
|
| 56 |
+
"valid": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/validation_clean.csv",
|
| 57 |
+
"test": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/test_clean.csv",
|
| 58 |
},
|
| 59 |
"mix": {
|
| 60 |
"train": "https://huggingface.co/datasets/semeru/completeformer-masked/resolve/main/training_mix.csv",
|