Datasets:
Fix encodings
Browse files- the_pile_github.py +8 -8
the_pile_github.py
CHANGED
@@ -18,6 +18,7 @@ import os
|
|
18 |
import re
|
19 |
import pandas as pd
|
20 |
import datasets
|
|
|
21 |
|
22 |
|
23 |
_CITATION = """\
|
@@ -135,7 +136,7 @@ class SmartContracts(datasets.GeneratorBasedBuilder):
|
|
135 |
features = datasets.Features(
|
136 |
{
|
137 |
"text": datasets.Value("string"),
|
138 |
-
"meta":
|
139 |
}
|
140 |
)
|
141 |
|
@@ -164,13 +165,13 @@ class SmartContracts(datasets.GeneratorBasedBuilder):
|
|
164 |
# By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
|
165 |
train_urls = _URLS["train"][self.config.name]
|
166 |
train_files = dl_manager.download_and_extract(train_urls)
|
167 |
-
|
168 |
-
|
169 |
|
170 |
return [
|
171 |
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"split_key": "train", "files": train_files}),
|
172 |
-
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"split_key": "validation", "files":
|
173 |
-
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"split_key": "test", "files":
|
174 |
]
|
175 |
|
176 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
@@ -179,7 +180,6 @@ class SmartContracts(datasets.GeneratorBasedBuilder):
|
|
179 |
# This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
|
180 |
# The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
|
181 |
#data = pd.read_parquet(filepath)
|
182 |
-
|
183 |
key = 0
|
184 |
for path in files:
|
185 |
data = pd.read_parquet(path)
|
@@ -189,7 +189,7 @@ class SmartContracts(datasets.GeneratorBasedBuilder):
|
|
189 |
if split_key == "train":
|
190 |
yield key, {
|
191 |
"text": row.text,
|
192 |
-
"meta": row.meta,
|
193 |
}
|
194 |
|
195 |
elif split_key in ["validation", "test"]:
|
@@ -199,7 +199,7 @@ class SmartContracts(datasets.GeneratorBasedBuilder):
|
|
199 |
"meta": row.meta,
|
200 |
}
|
201 |
else:
|
202 |
-
language = row.meta[
|
203 |
if language == self.config.name:
|
204 |
yield key, {
|
205 |
"text": row.text,
|
|
|
18 |
import re
|
19 |
import pandas as pd
|
20 |
import datasets
|
21 |
+
import json
|
22 |
|
23 |
|
24 |
_CITATION = """\
|
|
|
136 |
features = datasets.Features(
|
137 |
{
|
138 |
"text": datasets.Value("string"),
|
139 |
+
"meta": {'language': datasets.Value('string')},
|
140 |
}
|
141 |
)
|
142 |
|
|
|
165 |
# By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
|
166 |
train_urls = _URLS["train"][self.config.name]
|
167 |
train_files = dl_manager.download_and_extract(train_urls)
|
168 |
+
dev_file = dl_manager.download_and_extract(_URLS["dev"])
|
169 |
+
test_file = dl_manager.download_and_extract(_URLS["test"])
|
170 |
|
171 |
return [
|
172 |
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"split_key": "train", "files": train_files}),
|
173 |
+
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"split_key": "validation", "files": [dev_file]}),
|
174 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"split_key": "test", "files": [test_file]}),
|
175 |
]
|
176 |
|
177 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
|
|
180 |
# This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
|
181 |
# The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
|
182 |
#data = pd.read_parquet(filepath)
|
|
|
183 |
key = 0
|
184 |
for path in files:
|
185 |
data = pd.read_parquet(path)
|
|
|
189 |
if split_key == "train":
|
190 |
yield key, {
|
191 |
"text": row.text,
|
192 |
+
"meta": json.loads(row.meta.decode('utf-8')),
|
193 |
}
|
194 |
|
195 |
elif split_key in ["validation", "test"]:
|
|
|
199 |
"meta": row.meta,
|
200 |
}
|
201 |
else:
|
202 |
+
language = row.meta["language"].lower().replace(" ", "_") # e.g. "Jupyter Notebook" -> "jupyter_notebook"
|
203 |
if language == self.config.name:
|
204 |
yield key, {
|
205 |
"text": row.text,
|