system HF staff commited on
Commit
f843faa
1 Parent(s): 2f2796b

Update files from the datasets library (from 1.6.0)

Browse files

Release notes: https://github.com/huggingface/datasets/releases/tag/1.6.0

Files changed (1) hide show
  1. glue.py +7 -6
glue.py CHANGED
@@ -16,14 +16,12 @@
16
  # Lint as: python3
17
  """The General Language Understanding Evaluation (GLUE) benchmark."""
18
 
19
- from __future__ import absolute_import, division, print_function
20
 
21
  import csv
22
  import os
23
  import textwrap
24
 
25
  import numpy as np
26
- import six
27
 
28
  import datasets
29
 
@@ -453,7 +451,7 @@ class Glue(datasets.GeneratorBasedBuilder):
453
  ]
454
 
455
  def _info(self):
456
- features = {text_feature: datasets.Value("string") for text_feature in six.iterkeys(self.config.text_features)}
457
  if self.config.label_classes:
458
  features["label"] = datasets.features.ClassLabel(names=self.config.label_classes)
459
  else:
@@ -565,7 +563,7 @@ class Glue(datasets.GeneratorBasedBuilder):
565
  "is_acceptable": row[1],
566
  }
567
 
568
- example = {feat: row[col] for feat, col in six.iteritems(self.config.text_features)}
569
  example["idx"] = n
570
 
571
  if self.config.label_column in row:
@@ -579,7 +577,7 @@ class Glue(datasets.GeneratorBasedBuilder):
579
  example["label"] = process_label(-1)
580
 
581
  # Filter out corrupted rows.
582
- for value in six.itervalues(example):
583
  if value is None:
584
  break
585
  else:
@@ -588,12 +586,15 @@ class Glue(datasets.GeneratorBasedBuilder):
588
  def _generate_example_mrpc_files(self, mrpc_files, split):
589
  if split == "test":
590
  with open(mrpc_files["test"], encoding="utf8") as f:
 
 
 
591
  reader = csv.DictReader(f, delimiter="\t", quoting=csv.QUOTE_NONE)
592
  for n, row in enumerate(reader):
593
  yield {
594
  "sentence1": row["#1 String"],
595
  "sentence2": row["#2 String"],
596
- "label": -1,
597
  "idx": n,
598
  }
599
  else:
 
16
  # Lint as: python3
17
  """The General Language Understanding Evaluation (GLUE) benchmark."""
18
 
 
19
 
20
  import csv
21
  import os
22
  import textwrap
23
 
24
  import numpy as np
 
25
 
26
  import datasets
27
 
 
451
  ]
452
 
453
  def _info(self):
454
+ features = {text_feature: datasets.Value("string") for text_feature in self.config.text_features.keys()}
455
  if self.config.label_classes:
456
  features["label"] = datasets.features.ClassLabel(names=self.config.label_classes)
457
  else:
 
563
  "is_acceptable": row[1],
564
  }
565
 
566
+ example = {feat: row[col] for feat, col in self.config.text_features.items()}
567
  example["idx"] = n
568
 
569
  if self.config.label_column in row:
 
577
  example["label"] = process_label(-1)
578
 
579
  # Filter out corrupted rows.
580
+ for value in example.values():
581
  if value is None:
582
  break
583
  else:
 
586
  def _generate_example_mrpc_files(self, mrpc_files, split):
587
  if split == "test":
588
  with open(mrpc_files["test"], encoding="utf8") as f:
589
+ # The first 3 bytes are the utf-8 BOM \xef\xbb\xbf, which messes with
590
+ # the Quality key.
591
+ f.seek(3)
592
  reader = csv.DictReader(f, delimiter="\t", quoting=csv.QUOTE_NONE)
593
  for n, row in enumerate(reader):
594
  yield {
595
  "sentence1": row["#1 String"],
596
  "sentence2": row["#2 String"],
597
+ "label": int(row["Quality"]),
598
  "idx": n,
599
  }
600
  else: