nthngdy commited on
Commit
9938b56
1 Parent(s): d0f40ba

Update openwebtext_split.py

Browse files
Files changed (1) hide show
  1. openwebtext_split.py +5 -4
openwebtext_split.py CHANGED
@@ -19,7 +19,7 @@ import os
19
  import re
20
  from itertools import chain
21
  import tarfile
22
- import io
23
 
24
  import datasets
25
 
@@ -40,6 +40,7 @@ An open-source replication of the WebText dataset from OpenAI.
40
  _URL = "https://zenodo.org/record/3834942/files/openwebtext.tar.xz"
41
 
42
  def custom_iter_archive(path_or_buf, _filter=lambda x: True):
 
43
  def _iter_archive(f):
44
  stream = tarfile.open(fileobj=f, mode="r|*")
45
  for i, tarinfo in enumerate(stream):
@@ -56,10 +57,10 @@ def custom_iter_archive(path_or_buf, _filter=lambda x: True):
56
  if not file_path.endswith('xz'):
57
  continue
58
  file_obj = stream.extractfile(tarinfo)
59
- print(file_obj.getmembers())
 
60
  for j, xzinfo in enumerate(file_obj):
61
- f = io.BytesIO(xzinfo)
62
- yield f
63
  stream.members = []
64
  del stream
65
  if hasattr(path_or_buf, "read"):
 
19
  import re
20
  from itertools import chain
21
  import tarfile
22
+ import lzma
23
 
24
  import datasets
25
 
 
40
  _URL = "https://zenodo.org/record/3834942/files/openwebtext.tar.xz"
41
 
42
  def custom_iter_archive(path_or_buf, _filter=lambda x: True):
43
+ decompressor = lzma.LZMADecompressor()
44
  def _iter_archive(f):
45
  stream = tarfile.open(fileobj=f, mode="r|*")
46
  for i, tarinfo in enumerate(stream):
 
57
  if not file_path.endswith('xz'):
58
  continue
59
  file_obj = stream.extractfile(tarinfo)
60
+ decompressed_file = decompressor.decompress(file_obj)
61
+ print(type(decompressed_file))
62
  for j, xzinfo in enumerate(file_obj):
63
+ yield j
 
64
  stream.members = []
65
  del stream
66
  if hasattr(path_or_buf, "read"):