qurator-spk/ocrd_calamari
1
0
Fork 0
mirror of https://github.com/mikegerber/ocrd_calamari.git synced 2025-06-09 11:49:53 +02:00
Code Releases Activity

✅ Fix tests by 1. binarizing and 2. use the GT4HistOCR model

Browse source
This commit is contained in:
Gerber, Mike 2019-12-05 13:02:20 +01:00
parent 2aff9d8a48
commit 99d04ddccb
4 changed files with 27 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

15
test/test_recognize.py
View file

@ -1,5 +1,6 @@
import os
import shutil
import subprocess
import urllib.request
from test.base import TestCase, main, assets, skip
@ -31,6 +32,18 @@ class TestCalamariRecognize(TestCase):
"https://github.com/OCR-D/assets/raw/master/data/kant_aufklaerung_1784/data/OCR-D-IMG/" + f,
os.path.join(WORKSPACE_DIR, 'OCR-D-IMG', f))
# The binarization options I have are:
#
# a. ocrd_kraken which tries to install cltsm, whose installation is borken on my machine (protobuf)
# b. ocrd_olena which 1. I cannot fully install via pip and 2. whose dependency olena doesn't compile on my
# machine
# c. just fumble with the original files
#
# So I'm going for option c.
for f in ['INPUT_0017.tif', 'INPUT_0020.tif']:
ff = os.path.join(WORKSPACE_DIR, 'OCR-D-IMG', f)
subprocess.call(['convert', ff, '-colorspace', 'Gray', ff])
# XXX Should remove GT text to really test this
CalamariRecognize(
@ -38,7 +51,7 @@ class TestCalamariRecognize(TestCase):
input_file_grp="OCR-D-GT-SEG-LINE",
output_file_grp="OCR-D-OCR-CALAMARI",
parameter={
'checkpoint': os.path.join(os.getcwd(), 'calamari_models/fraktur_19th_century/*.ckpt.json')
'checkpoint': os.path.join(os.getcwd(), 'gt4histocr-calamari/*.ckpt.json')
}
).process()
workspace.save_mets()