summaryrefslogtreecommitdiff
path: root/text_recognizer/data/sentence_generator.py
diff options
context:
space:
mode:
authorGustaf Rydholm <gustaf.rydholm@gmail.com>2021-03-28 22:02:24 +0200
committerGustaf Rydholm <gustaf.rydholm@gmail.com>2021-03-28 22:02:24 +0200
commit46a1472d33d3a4180798492e819f2ec02bc3b1a3 (patch)
tree22322ed0d8f9f803966ea745ec5bb8c759f8db64 /text_recognizer/data/sentence_generator.py
parent8248f173132dfb7e47ec62b08e9235990c8626e3 (diff)
Add refactor of iam lines
Diffstat (limited to 'text_recognizer/data/sentence_generator.py')
-rw-r--r--text_recognizer/data/sentence_generator.py7
1 files changed, 3 insertions, 4 deletions
diff --git a/text_recognizer/data/sentence_generator.py b/text_recognizer/data/sentence_generator.py
index 53b781c..f09703b 100644
--- a/text_recognizer/data/sentence_generator.py
+++ b/text_recognizer/data/sentence_generator.py
@@ -1,5 +1,4 @@
"""Downloading the Brown corpus with NLTK for sentence generating."""
-
import itertools
import re
import string
@@ -9,9 +8,9 @@ import nltk
from nltk.corpus.reader.util import ConcatenatedCorpusView
import numpy as np
-from text_recognizer.datasets.util import DATA_DIRNAME
+from text_recognizer.data.base_data_module import BaseDataModule
-NLTK_DATA_DIRNAME = DATA_DIRNAME / "downloaded" / "nltk"
+NLTK_DATA_DIRNAME = BaseDataModule.data_dirname() / "downloaded" / "nltk"
class SentenceGenerator:
@@ -47,7 +46,7 @@ class SentenceGenerator:
raise ValueError(
"Must provide max_length to this method or when making this object."
)
-
+
for _ in range(10):
try:
index = np.random.randint(0, len(self.word_start_indices) - 1)