rain1024 commited on
Commit
e192c17
1 Parent(s): 96432bc
Files changed (4) hide show
  1. UTS_Text.py +2 -4
  2. generate_dataset.py +56 -1
  3. statistics.py +5 -2
  4. test_dataset.py +3 -3
UTS_Text.py CHANGED
@@ -1,6 +1,4 @@
1
  import os
2
- import re
3
- from itertools import chain
4
 
5
  import datasets
6
 
@@ -23,6 +21,7 @@ class UTSTextConfig(datasets.BuilderConfig):
23
  def __init__(self, **kwargs):
24
  super(UTSTextConfig, self).__init__(**kwargs)
25
 
 
26
  class UTSText(datasets.GeneratorBasedBuilder):
27
  """UTS Word Tokenize datasets"""
28
  VERSION = datasets.Version("1.0.0")
@@ -47,7 +46,7 @@ class UTSText(datasets.GeneratorBasedBuilder):
47
  homepage=None,
48
  citation=_CITATION
49
  )
50
-
51
  def _split_generators(self, dl_manager):
52
  """Returns SplitGenerators."""
53
  subset_folder = self.config.name
@@ -83,4 +82,3 @@ class UTSText(datasets.GeneratorBasedBuilder):
83
  }
84
  yield guid, item
85
  guid += 1
86
-
 
1
  import os
 
 
2
 
3
  import datasets
4
 
 
21
  def __init__(self, **kwargs):
22
  super(UTSTextConfig, self).__init__(**kwargs)
23
 
24
+
25
  class UTSText(datasets.GeneratorBasedBuilder):
26
  """UTS Word Tokenize datasets"""
27
  VERSION = datasets.Version("1.0.0")
 
46
  homepage=None,
47
  citation=_CITATION
48
  )
49
+
50
  def _split_generators(self, dl_manager):
51
  """Returns SplitGenerators."""
52
  subset_folder = self.config.name
 
82
  }
83
  yield guid, item
84
  guid += 1
 
generate_dataset.py CHANGED
@@ -1 +1,56 @@
1
- from underthesea.transformer.tagged_feature import lower_words
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from os.path import join, dirname
2
+ from underthesea.file_utils import DATASETS_FOLDER
3
+ import random
4
+
5
+ random.seed(10)
6
+ text_file = join(DATASETS_FOLDER, "VNESES", "VNESEScorpus.txt")
7
+ with open(text_file) as f:
8
+ lines = f.read().splitlines()
9
+ NUM_LONG_TOKENS = 50
10
+ NUM_SHORT_TOKENS = 20
11
+ NUM_MAX_TOKENS = 200
12
+
13
+
14
+ def longline_conditions(line):
15
+ if len(line) < NUM_LONG_TOKENS or len(line) > NUM_MAX_TOKENS:
16
+ return False
17
+ if not (line[0].isupper() and line[-1] == "."):
18
+ return False
19
+ return True
20
+
21
+
22
+ long_lines = [line for line in lines if longline_conditions(line)]
23
+ print("Short lines", len(long_lines))
24
+
25
+
26
+ def shortline_conditions(line):
27
+ if len(line) < NUM_SHORT_TOKENS:
28
+ return False
29
+ if len(line) > NUM_LONG_TOKENS:
30
+ return False
31
+ if not line[0].isupper():
32
+ return False
33
+ return True
34
+
35
+
36
+ short_lines = [line for line in lines if shortline_conditions(line)]
37
+ print("Short lines", len(short_lines))
38
+
39
+ print("Long lines", len(long_lines))
40
+ print("Short lines", len(short_lines))
41
+
42
+ pwd = dirname(__file__)
43
+ tmp = join(pwd, "data")
44
+ corpus_file = join(tmp, "UTS_Text_v1.txt")
45
+
46
+ with open(corpus_file, "w") as f:
47
+ lines = long_lines + short_lines
48
+ lines = lines[:1000]
49
+ content = "\n".join(lines)
50
+ f.write(content)
51
+
52
+ with open(join(tmp, "large", "train.txt"), "w") as f:
53
+ lines = long_lines + short_lines
54
+ lines = lines[:100000]
55
+ content = "\n".join(lines)
56
+ f.write(content)
statistics.py CHANGED
@@ -1,17 +1,20 @@
1
  from os.path import dirname, join
2
 
 
3
  def count_lines(filepath):
4
  with open(filepath, encoding="utf-8") as f:
5
  lines = f.read().splitlines()
6
  return len(lines)
7
 
 
8
  def statistic(subset_folder):
9
  print("Train\t\t", count_lines(join(subset_folder, "train.txt")))
10
  print("Validation\t", count_lines(join(subset_folder, "validation.txt")))
11
  print("Test\t\t", count_lines(join(subset_folder, "test.txt")))
12
 
 
13
  pwd = dirname(__file__)
14
- dataset_folder = join(pwd, "data")
15
 
16
  print("Subset Small")
17
  statistic(join(dataset_folder, "small"))
@@ -20,4 +23,4 @@ print("Subset Base")
20
  statistic(join(dataset_folder, "base"))
21
 
22
  print("Subset Large")
23
- statistic(join(dataset_folder, "large"))
 
1
  from os.path import dirname, join
2
 
3
+
4
  def count_lines(filepath):
5
  with open(filepath, encoding="utf-8") as f:
6
  lines = f.read().splitlines()
7
  return len(lines)
8
 
9
+
10
  def statistic(subset_folder):
11
  print("Train\t\t", count_lines(join(subset_folder, "train.txt")))
12
  print("Validation\t", count_lines(join(subset_folder, "validation.txt")))
13
  print("Test\t\t", count_lines(join(subset_folder, "test.txt")))
14
 
15
+
16
  pwd = dirname(__file__)
17
+ dataset_folder = join(pwd, "data")
18
 
19
  print("Subset Small")
20
  statistic(join(dataset_folder, "small"))
 
23
  statistic(join(dataset_folder, "base"))
24
 
25
  print("Subset Large")
26
+ statistic(join(dataset_folder, "large"))
test_dataset.py CHANGED
@@ -1,6 +1,6 @@
1
  from datasets import load_dataset
2
 
3
 
4
- dataset = load_dataset("undertheseanlp/UTS_Text", "small")
5
-
6
- print(dataset["train"][0])
 
1
  from datasets import load_dataset
2
 
3
 
4
+ if __name__ == '__main__':
5
+ dataset = load_dataset("undertheseanlp/UTS_Text", "small")
6
+ print(dataset["train"][0])