Datasets:
Tasks:
Text Generation
Sub-tasks:
language-modeling
Languages:
code
Multilinguality:
multilingual
Size Categories:
unknown
Language Creators:
crowdsourced
Annotations Creators:
no-annotation
Source Datasets:
original
ArXiv:
License:
ncoop57
commited on
Commit
•
ff0b80a
1
Parent(s):
4ba9fe8
Fix bug where I overwrite datasets import
Browse files- code_clippy.py +2 -2
code_clippy.py
CHANGED
|
@@ -50,12 +50,12 @@ _HOMEPAGE = "https://github.com/CodedotAl/gpt-code-clippy/wiki"
|
|
| 50 |
|
| 51 |
_LICENSE = "GPL-3.0"
|
| 52 |
|
| 53 |
-
|
| 54 |
splits = ["train", "validation", "test"]
|
| 55 |
|
| 56 |
BASE_URL = "https://the-eye.eu/public/AI/training_data/code_clippy_data/"
|
| 57 |
_URLs = {}
|
| 58 |
-
for dataset in
|
| 59 |
_URLs[dataset] = {}
|
| 60 |
for split in splits:
|
| 61 |
_URLs[dataset][split] = []
|
|
|
|
| 50 |
|
| 51 |
_LICENSE = "GPL-3.0"
|
| 52 |
|
| 53 |
+
dataset_names = ["code_clippy_dedup_data", "code_clippy_dup_data"]
|
| 54 |
splits = ["train", "validation", "test"]
|
| 55 |
|
| 56 |
BASE_URL = "https://the-eye.eu/public/AI/training_data/code_clippy_data/"
|
| 57 |
_URLs = {}
|
| 58 |
+
for dataset in dataset_names:
|
| 59 |
_URLs[dataset] = {}
|
| 60 |
for split in splits:
|
| 61 |
_URLs[dataset][split] = []
|