Upload loaders.py with huggingface_hub
Browse files- loaders.py +99 -15
loaders.py
CHANGED
@@ -34,6 +34,7 @@ import pandas as pd
|
|
34 |
from datasets import load_dataset as hf_load_dataset
|
35 |
from tqdm import tqdm
|
36 |
|
|
|
37 |
from .logging_utils import get_logger
|
38 |
from .operator import SourceOperator
|
39 |
from .settings_utils import get_settings
|
@@ -45,8 +46,6 @@ settings = get_settings()
|
|
45 |
try:
|
46 |
import ibm_boto3
|
47 |
|
48 |
-
# from ibm_botocore.client import ClientError
|
49 |
-
|
50 |
ibm_boto3_available = True
|
51 |
except ImportError:
|
52 |
ibm_boto3_available = False
|
@@ -62,6 +61,27 @@ class Loader(SourceOperator):
|
|
62 |
loader_limit: int = None
|
63 |
streaming: bool = False
|
64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
|
66 |
class LoadHF(Loader):
|
67 |
path: str
|
@@ -71,10 +91,11 @@ class LoadHF(Loader):
|
|
71 |
data_files: Optional[
|
72 |
Union[str, Sequence[str], Mapping[str, Union[str, Sequence[str]]]]
|
73 |
] = None
|
74 |
-
streaming: bool =
|
|
|
75 |
|
76 |
-
def
|
77 |
-
|
78 |
with tempfile.TemporaryDirectory() as dir_to_be_deleted:
|
79 |
try:
|
80 |
dataset = hf_load_dataset(
|
@@ -92,11 +113,18 @@ class LoadHF(Loader):
|
|
92 |
raise ValueError(
|
93 |
f"{self.__class__.__name__} cannot run remote code from huggingface without setting unitxt.settings.allow_unverified_code=True or by setting environment vairable: UNITXT_ALLOW_UNVERIFIED_CODE."
|
94 |
) from e
|
|
|
95 |
if self.split is not None:
|
96 |
dataset = {self.split: dataset}
|
97 |
-
|
98 |
-
|
99 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
100 |
with tempfile.TemporaryDirectory() as dir_to_be_deleted:
|
101 |
try:
|
102 |
dataset = hf_load_dataset(
|
@@ -121,17 +149,73 @@ class LoadHF(Loader):
|
|
121 |
else:
|
122 |
dataset = {self.split: dataset}
|
123 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
124 |
return MultiStream.from_iterables(dataset)
|
125 |
|
126 |
|
127 |
class LoadCSV(Loader):
|
128 |
files: Dict[str, str]
|
129 |
chunksize: int = 1000
|
|
|
|
|
|
|
130 |
|
131 |
def stream_csv(self, file):
|
132 |
-
|
133 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
134 |
yield row.to_dict()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
135 |
|
136 |
def process(self):
|
137 |
if self.streaming:
|
@@ -144,7 +228,7 @@ class LoadCSV(Loader):
|
|
144 |
|
145 |
return MultiStream(
|
146 |
{
|
147 |
-
name:
|
148 |
for name, file in self.files.items()
|
149 |
}
|
150 |
)
|
@@ -211,17 +295,17 @@ class LoadFromIBMCloud(Loader):
|
|
211 |
f"Unabled to access {item_name} in {bucket_name} in COS", e
|
212 |
) from e
|
213 |
|
214 |
-
if self.
|
215 |
if item_name.endswith(".jsonl"):
|
216 |
first_lines = list(
|
217 |
-
itertools.islice(body.iter_lines(), self.
|
218 |
)
|
219 |
with open(local_file, "wb") as downloaded_file:
|
220 |
for line in first_lines:
|
221 |
downloaded_file.write(line)
|
222 |
downloaded_file.write(b"\n")
|
223 |
logger.info(
|
224 |
-
f"\nDownload successful limited to {self.
|
225 |
)
|
226 |
return
|
227 |
|
@@ -277,7 +361,7 @@ class LoadFromIBMCloud(Loader):
|
|
277 |
self.cache_dir,
|
278 |
self.bucket_name,
|
279 |
self.data_dir,
|
280 |
-
f"loader_limit_{self.
|
281 |
)
|
282 |
if not os.path.exists(local_dir):
|
283 |
Path(local_dir).mkdir(parents=True, exist_ok=True)
|
|
|
34 |
from datasets import load_dataset as hf_load_dataset
|
35 |
from tqdm import tqdm
|
36 |
|
37 |
+
from .dataclass import InternalField
|
38 |
from .logging_utils import get_logger
|
39 |
from .operator import SourceOperator
|
40 |
from .settings_utils import get_settings
|
|
|
46 |
try:
|
47 |
import ibm_boto3
|
48 |
|
|
|
|
|
49 |
ibm_boto3_available = True
|
50 |
except ImportError:
|
51 |
ibm_boto3_available = False
|
|
|
61 |
loader_limit: int = None
|
62 |
streaming: bool = False
|
63 |
|
64 |
+
def get_limit(self):
|
65 |
+
if settings.global_loader_limit is not None and self.loader_limit is not None:
|
66 |
+
return min(int(settings.global_loader_limit), self.loader_limit)
|
67 |
+
if settings.global_loader_limit is not None:
|
68 |
+
return int(settings.global_loader_limit)
|
69 |
+
return self.loader_limit
|
70 |
+
|
71 |
+
def get_limiter(self):
|
72 |
+
if settings.global_loader_limit is not None and self.loader_limit is not None:
|
73 |
+
if int(settings.global_loader_limit) > self.loader_limit:
|
74 |
+
return f"{self.__class__.__name__}.loader_limit"
|
75 |
+
return "unitxt.settings.global_loader_limit"
|
76 |
+
if settings.global_loader_limit is not None:
|
77 |
+
return "unitxt.settings.global_loader_limit"
|
78 |
+
return f"{self.__class__.__name__}.loader_limit"
|
79 |
+
|
80 |
+
def log_limited_loading(self):
|
81 |
+
logger.info(
|
82 |
+
f"\nLoading limited to {self.get_limit()} instances by setting {self.get_limiter()};"
|
83 |
+
)
|
84 |
+
|
85 |
|
86 |
class LoadHF(Loader):
|
87 |
path: str
|
|
|
91 |
data_files: Optional[
|
92 |
Union[str, Sequence[str], Mapping[str, Union[str, Sequence[str]]]]
|
93 |
] = None
|
94 |
+
streaming: bool = True
|
95 |
+
_cache: dict = InternalField(default=None)
|
96 |
|
97 |
+
def stream_dataset(self):
|
98 |
+
if self._cache is None:
|
99 |
with tempfile.TemporaryDirectory() as dir_to_be_deleted:
|
100 |
try:
|
101 |
dataset = hf_load_dataset(
|
|
|
113 |
raise ValueError(
|
114 |
f"{self.__class__.__name__} cannot run remote code from huggingface without setting unitxt.settings.allow_unverified_code=True or by setting environment vairable: UNITXT_ALLOW_UNVERIFIED_CODE."
|
115 |
) from e
|
116 |
+
|
117 |
if self.split is not None:
|
118 |
dataset = {self.split: dataset}
|
119 |
+
|
120 |
+
self._cache = dataset
|
121 |
+
else:
|
122 |
+
dataset = self._cache
|
123 |
+
|
124 |
+
return dataset
|
125 |
+
|
126 |
+
def load_dataset(self):
|
127 |
+
if self._cache is None:
|
128 |
with tempfile.TemporaryDirectory() as dir_to_be_deleted:
|
129 |
try:
|
130 |
dataset = hf_load_dataset(
|
|
|
149 |
else:
|
150 |
dataset = {self.split: dataset}
|
151 |
|
152 |
+
self._cache = dataset
|
153 |
+
else:
|
154 |
+
dataset = self._cache
|
155 |
+
|
156 |
+
return dataset
|
157 |
+
|
158 |
+
def split_limited_load(self, split_name):
|
159 |
+
yield from itertools.islice(self._cache[split_name], self.get_limit())
|
160 |
+
|
161 |
+
def limited_load(self):
|
162 |
+
self.log_limited_loading()
|
163 |
+
return MultiStream(
|
164 |
+
{
|
165 |
+
name: Stream(
|
166 |
+
generator=self.split_limited_load, gen_kwargs={"split_name": name}
|
167 |
+
)
|
168 |
+
for name in self._cache.keys()
|
169 |
+
}
|
170 |
+
)
|
171 |
+
|
172 |
+
def process(self):
|
173 |
+
try:
|
174 |
+
dataset = self.stream_dataset()
|
175 |
+
except (
|
176 |
+
NotImplementedError
|
177 |
+
): # streaming is not supported for zipped files so we load without streaming
|
178 |
+
dataset = self.load_dataset()
|
179 |
+
|
180 |
+
if self.get_limit() is not None:
|
181 |
+
return self.limited_load()
|
182 |
+
|
183 |
return MultiStream.from_iterables(dataset)
|
184 |
|
185 |
|
186 |
class LoadCSV(Loader):
|
187 |
files: Dict[str, str]
|
188 |
chunksize: int = 1000
|
189 |
+
_cache = InternalField(default_factory=dict)
|
190 |
+
loader_limit: int = None
|
191 |
+
streaming: bool = True
|
192 |
|
193 |
def stream_csv(self, file):
|
194 |
+
if self.get_limit() is not None:
|
195 |
+
self.log_limited_loading()
|
196 |
+
chunksize = min(self.get_limit(), self.chunksize)
|
197 |
+
else:
|
198 |
+
chunksize = self.chunksize
|
199 |
+
|
200 |
+
row_count = 0
|
201 |
+
for chunk in pd.read_csv(file, chunksize=chunksize):
|
202 |
+
for _, row in chunk.iterrows():
|
203 |
+
if self.get_limit() is not None and row_count >= self.get_limit():
|
204 |
+
return
|
205 |
yield row.to_dict()
|
206 |
+
row_count += 1
|
207 |
+
|
208 |
+
def load_csv(self, file):
|
209 |
+
if file not in self._cache:
|
210 |
+
if self.get_limit() is not None:
|
211 |
+
self.log_limited_loading()
|
212 |
+
self._cache[file] = pd.read_csv(file, nrows=self.get_limit()).to_dict(
|
213 |
+
"records"
|
214 |
+
)
|
215 |
+
else:
|
216 |
+
self._cache[file] = pd.read_csv(file).to_dict("records")
|
217 |
+
|
218 |
+
yield from self._cache[file]
|
219 |
|
220 |
def process(self):
|
221 |
if self.streaming:
|
|
|
228 |
|
229 |
return MultiStream(
|
230 |
{
|
231 |
+
name: Stream(generator=self.load_csv, gen_kwargs={"file": file})
|
232 |
for name, file in self.files.items()
|
233 |
}
|
234 |
)
|
|
|
295 |
f"Unabled to access {item_name} in {bucket_name} in COS", e
|
296 |
) from e
|
297 |
|
298 |
+
if self.get_limit() is not None:
|
299 |
if item_name.endswith(".jsonl"):
|
300 |
first_lines = list(
|
301 |
+
itertools.islice(body.iter_lines(), self.get_limit())
|
302 |
)
|
303 |
with open(local_file, "wb") as downloaded_file:
|
304 |
for line in first_lines:
|
305 |
downloaded_file.write(line)
|
306 |
downloaded_file.write(b"\n")
|
307 |
logger.info(
|
308 |
+
f"\nDownload successful limited to {self.get_limit()} lines"
|
309 |
)
|
310 |
return
|
311 |
|
|
|
361 |
self.cache_dir,
|
362 |
self.bucket_name,
|
363 |
self.data_dir,
|
364 |
+
f"loader_limit_{self.get_limit()}",
|
365 |
)
|
366 |
if not os.path.exists(local_dir):
|
367 |
Path(local_dir).mkdir(parents=True, exist_ok=True)
|