Switch URLs from MediaFire to GitHub Releases.

faustomorales · Sep 13, 2020 · c8a4137 · c8a4137
1 parent 6b11251
commit c8a4137
Show file tree

Hide file tree

Showing 8 changed files with 45 additions and 30 deletions.
diff --git a/Dockerfile b/Dockerfile
@@ -4,8 +4,8 @@ WORKDIR /usr/src
 # change.
 RUN mkdir -p /root/.keras-ocr && ( \
     cd /root/.keras-ocr && \
-    curl -L -o craft_mlt_25k.h5 https://www.mediafire.com/file/mepzf3sq7u7nve9/craft_mlt_25k.h5/file && \
-    curl -L -o crnn_kurapan.h5 https://www.mediafire.com/file/pkj2p29b1f6fpil/crnn_kurapan.h5/file \
+    curl -L -o craft_mlt_25k.h5 https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/craft_mlt_25k.h5 && \
+    curl -L -o crnn_kurapan.h5 https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/crnn_kurapan.h5 \
     )
 COPY ./Pipfile* ./
 COPY ./Makefile ./

diff --git a/README.md b/README.md
@@ -61,10 +61,10 @@ You may be wondering how the models in this package compare to existing cloud OC
 
 | model                 | latency | precision | recall |
 |-----------------------|---------|-----------|--------|
-| [AWS](https://www.mediafire.com/file/7obsgyzg7z1ltb0/aws_annotations.json/file)                   | 719ms   | 0.45      | 0.48   |
-| [GCP](https://www.mediafire.com/file/8is5pq161ui95ox/google_annotations.json/file)                   | 388ms   | 0.53      | 0.58   |
-| [keras-ocr](https://www.mediafire.com/file/1gcwtrzy537v0sn/keras_ocr_annotations_scale_2.json/file) (scale=2)  | 417ms   | 0.53      | 0.54   |
-| [keras-ocr](https://www.mediafire.com/file/dc7e66oupelsp7p/keras_ocr_annotations_scale_3.json/file) (scale=3)  | 699ms   | 0.5       | 0.59   |
+| [AWS](https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/aws_annotations.json)                   | 719ms   | 0.45      | 0.48   |
+| [GCP](https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/google_annotations.json)                   | 388ms   | 0.53      | 0.58   |
+| [keras-ocr](https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/keras_ocr_annotations_scale_2.json) (scale=2)  | 417ms   | 0.53      | 0.54   |
+| [keras-ocr](https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/keras_ocr_annotations_scale_3.json) (scale=3)  | 699ms   | 0.5       | 0.59   |
 
 - Precision and recall were computed based on an intersection over union of 50% or higher and a text similarity to ground truth of 50% or higher.
 - `keras-ocr` latency values were computed using a Tesla P4 GPU on Google Colab. `scale` refers to the argument provided to `keras_ocr.pipelines.Pipeline()` which determines the upscaling applied to the image prior to inference.

diff --git a/keras_ocr/data_generation.py b/keras_ocr/data_generation.py
@@ -1,4 +1,4 @@
-# pylint: disable=invalid-name,too-many-locals,too-many-arguments,too-many-branches,too-many-statements,stop-iteration-return
+# pylint: disable=invalid-name,line-too-long,too-many-locals,too-many-arguments,too-many-branches,too-many-statements,stop-iteration-return
 import os
 import math
 import glob
@@ -168,7 +168,7 @@ def get_backgrounds(cache_dir=None):
         cache_dir = os.path.expanduser(os.path.join('~', '.keras-ocr'))
     backgrounds_dir = os.path.join(cache_dir, 'backgrounds')
     backgrounds_zip_path = tools.download_and_verify(
-        url='https://www.mediafire.com/file/l0pdx5j860kqmyr/backgrounds.zip/file',
+        url='https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/backgrounds.zip',
         sha256='f263ed0d55de303185cc0f93e9fcb0b13104d68ed71af7aaaa8e8c91389db471',
         filename='backgrounds.zip',
         cache_dir=cache_dir)
@@ -199,7 +199,7 @@ def get_fonts(cache_dir=None,
     if cache_dir is None:
         cache_dir = os.path.expanduser(os.path.join('~', '.keras-ocr'))
     fonts_zip_path = tools.download_and_verify(
-        url='https://www.mediafire.com/file/6v9r9oztyri0jrc/fonts.zip/file',
+        url='https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/fonts.zip',
         sha256='d4d90c27a9bc4bf8fff1d2c0a00cfb174c7d5d10f60ed29d5f149ef04d45b700',
         filename='fonts.zip',
         cache_dir=cache_dir)
@@ -212,7 +212,8 @@ def get_fonts(cache_dir=None,
     if exclude_smallcaps:
         with open(
                 tools.download_and_verify(
-                    url='https://www.mediafire.com/file/v2o7hxn0mapne7i/fonts_smallcaps.txt/file',
+                    url=
+                    'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/fonts_smallcaps.txt',
                     sha256='6531c700523c687f02852087530d1ab3c7cc0b59891bbecc77726fbb0aabe68e',
                     filename='fonts_smallcaps.txt',
                     cache_dir=cache_dir), 'r') as f:
@@ -312,7 +313,6 @@ def draw_text_image(text,
         box is an array of points with shape (4, 2) providing the coordinates
         of the character box in clockwise order starting from the top left.
     """
-    # pylint: disable=bad-continuation
     if not use_ligatures:
         fonts = {
             subalphabet: PIL.ImageFont.truetype(font_path, size=fontsize)

diff --git a/keras_ocr/datasets.py b/keras_ocr/datasets.py
@@ -1,6 +1,7 @@
-# pylint: disable=invalid-name,too-many-arguments,too-many-locals
+# pylint: disable=line-too-long,invalid-name,too-many-arguments,too-many-locals
 import concurrent
 import itertools
+import warnings
 import zipfile
 import random
 import glob
@@ -120,7 +121,7 @@ def get_born_digital_recognizer_dataset(split='train', cache_dir=None):
         train_dir = os.path.join(main_dir, 'train')
         training_zip_path = tools.download_and_verify(
             url=
-            'https://www.mediafire.com/file/ybj0uo196rushhn/Challenge1_Training_Task3_Images_GT.zip/file',  # pylint: disable=line-too-long
+            'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/Challenge1_Training_Task3_Images_GT.zip',  # pylint: disable=line-too-long
             filename='Challenge1_Training_Task3_Images_GT.zip',
             cache_dir=main_dir,
             sha256='8ede0639f5a8031d584afd98cee893d1c5275d7f17863afc2cba24b13c932b07')
@@ -136,15 +137,16 @@ def get_born_digital_recognizer_dataset(split='train', cache_dir=None):
         test_dir = os.path.join(main_dir, 'test')
         test_zip_path = tools.download_and_verify(
             url=
-            'https://www.mediafire.com/file/nesckvjulvzpb2i/Challenge1_Test_Task3_Images.zip/file',
+            'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/Challenge1_Test_Task3_Images.zip',
             filename='Challenge1_Test_Task3_Images.zip',
             cache_dir=main_dir,
             sha256='8f781b0140fd0bac3750530f0924bce5db3341fd314a2fcbe9e0b6ca409a77f0')
         if len(glob.glob(os.path.join(test_dir, '*.png'))) != 1439:
             with zipfile.ZipFile(test_zip_path) as zfile:
                 zfile.extractall(test_dir)
         test_gt_path = tools.download_and_verify(
-            url='https://www.mediafire.com/file/euuuwsgg7z4pcb2/Challenge1_Test_Task3_GT.txt/file',
+            url=
+            'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/Challenge1_Test_Task3_GT.txt',
             cache_dir=test_dir,
             filename='Challenge1_Test_Task3_GT.txt',
             sha256='fce7f1228b7c4c26a59f13f562085148acf063d6690ce51afc395e0a1aabf8be')
@@ -194,7 +196,7 @@ def get_icdar_2013_detector_dataset(cache_dir=None, skip_illegible=False):
     training_images_dir = os.path.join(main_dir, 'Challenge2_Training_Task12_Images')
     training_zip_images_path = tools.download_and_verify(
         url=
-        'https://www.mediafire.com/file/l8ct7ckudg12ln6/Challenge2_Training_Task12_Images.zip/file',  # pylint: disable=line-too-long
+        'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/Challenge2_Training_Task12_Images.zip',  # pylint: disable=line-too-long
         cache_dir=main_dir,
         filename='Challenge2_Training_Task12_Images.zip',
         sha256='7a57d1699fbb92db3ad82c930202938562edaf72e1c422ddd923860d8ace8ded')
@@ -203,7 +205,8 @@ def get_icdar_2013_detector_dataset(cache_dir=None, skip_illegible=False):
             zfile.extractall(training_images_dir)
     training_gt_dir = os.path.join(main_dir, 'Challenge2_Training_Task2_GT')
     training_zip_gt_path = tools.download_and_verify(
-        url='https://www.mediafire.com/file/rpfphmxvudn5v3y/Challenge2_Training_Task2_GT.zip/file',  # pylint: disable=line-too-long
+        url=
+        'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/Challenge2_Training_Task2_GT.zip',  # pylint: disable=line-too-long
         cache_dir=main_dir,
         filename='Challenge2_Training_Task2_GT.zip',
         sha256='4cedd5b1e33dc4354058f5967221ac85dbdf91a99b30f3ab1ecdf42786a9d027')
@@ -245,6 +248,14 @@ def get_icdar_2019_semisupervised_dataset(cache_dir=None):
     Args:
         cache_dir: The cache directory to use.
     """
+    warnings.warn(
+        "You may need to get this dataset manually in-browser by downloading "
+        "https://www.mediafire.com/file/snekaezeextc3ee/ImagesPart1.zip/file "
+        "and https://www.mediafire.com/file/i2snljkfm4t2ojm/ImagesPart2.zip/file "
+        "and putting them in ~/.keras-ocr/icdar2019. The files are too big "
+        "for GitHub Releases and we may run out of direct download  bandwidth on "
+        "MediaFire where they are hosted. See "
+        "https://github.com/faustomorales/keras-ocr/issues/117 for more details.", UserWarning)
     if cache_dir is None:
         cache_dir = tools.get_default_cache_dir()
     main_dir = os.path.join(cache_dir, 'icdar2019')
@@ -267,7 +278,8 @@ def get_icdar_2019_semisupervised_dataset(cache_dir=None):
         with zipfile.ZipFile(training_zip_2) as zfile:
             zfile.extractall(main_dir)
     ground_truth = tools.download_and_verify(
-        url='http://www.mediafire.com/file/jshjv9kntxjzhva/mlt2019_dataset.json/file',  # pylint: disable=line-too-long
+        url=
+        'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/mlt2019_dataset.json',  # pylint: disable=line-too-long
         cache_dir=main_dir,
         filename='mlt2019_dataset.json')
     with open(ground_truth, 'r') as f:

diff --git a/keras_ocr/detection.py b/keras_ocr/detection.py
@@ -1,4 +1,4 @@
-# pylint: disable=invalid-name,too-many-locals,no-else-raise,too-many-arguments,no-self-use,too-many-statements,stop-iteration-return,import-outside-toplevel
+# pylint: disable=invalid-name,too-many-locals,line-too-long,no-else-raise,too-many-arguments,no-self-use,too-many-statements,stop-iteration-return,import-outside-toplevel
 import typing
 
 # The PyTorch portions of this code are subject to the following copyright notice.
@@ -426,7 +426,7 @@ def init_weights(modules):
 
     class vgg16_bn(torch.nn.Module):
         def __init__(self, pretrained=True, freeze=True):
-            super(vgg16_bn, self).__init__()
+            super().__init__()
             # We don't bother loading the pretrained VGG
             # because we're going to use the weights
             # at weights_path.
@@ -481,7 +481,7 @@ def forward(self, X):  # pylint: disable=arguments-differ
 
     class double_conv(nn.Module):
         def __init__(self, in_ch, mid_ch, out_ch):
-            super(double_conv, self).__init__()
+            super().__init__()
             self.conv = nn.Sequential(nn.Conv2d(in_ch + mid_ch, mid_ch, kernel_size=1),
                                       nn.BatchNorm2d(mid_ch), nn.ReLU(inplace=True),
                                       nn.Conv2d(mid_ch, out_ch, kernel_size=3, padding=1),
@@ -493,7 +493,7 @@ def forward(self, x):  # pylint: disable=arguments-differ
 
     class CRAFT(nn.Module):
         def __init__(self, pretrained=False, freeze=False):
-            super(CRAFT, self).__init__()
+            super().__init__()
             # Base network
             self.basenet = vgg16_bn(pretrained, freeze)
             # U network
@@ -567,12 +567,14 @@ def copyStateDict(state_dict):
 
 PRETRAINED_WEIGHTS = {
     ('clovaai_general', True): {
-        'url': 'https://www.mediafire.com/file/qh2ullnnywi320s/craft_mlt_25k.pth/file',
+        'url':
+        'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/craft_mlt_25k.pth',
         'filename': 'craft_mlt_25k.pth',
         'sha256': '4a5efbfb48b4081100544e75e1e2b57f8de3d84f213004b14b85fd4b3748db17'
     },
     ('clovaai_general', False): {
-        'url': 'https://www.mediafire.com/file/mepzf3sq7u7nve9/craft_mlt_25k.h5/file',
+        'url':
+        'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/craft_mlt_25k.h5',
         'filename': 'craft_mlt_25k.h5',
         'sha256': '7283ce2ff05a0617e9740c316175ff3bacdd7215dbdf1a726890d5099431f899'
     }

diff --git a/keras_ocr/recognition.py b/keras_ocr/recognition.py
@@ -1,4 +1,4 @@
-# pylint: disable=invalid-name,too-many-locals,too-many-arguments
+# pylint: disable=invalid-name,too-many-locals,too-many-arguments,line-too-long
 import typing
 import string
 
@@ -29,12 +29,14 @@
         'build_params': DEFAULT_BUILD_PARAMS,
         'weights': {
             'notop': {
-                'url': 'https://www.mediafire.com/file/n9yfn5wueu82rgf/crnn_kurapan_notop.h5/file',
+                'url':
+                'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/crnn_kurapan_notop.h5',
                 'filename': 'crnn_kurapan_notop.h5',
                 'sha256': '027fd2cced3cbea0c4f5894bb8e9e85bac04f11daf96b8fdcf1e4ee95dcf51b9'
             },
             'top': {
-                'url': 'https://www.mediafire.com/file/pkj2p29b1f6fpil/crnn_kurapan.h5/file',
+                'url':
+                'https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/crnn_kurapan.h5',
                 'filename': 'crnn_kurapan.h5',
                 'sha256': 'a7d8086ac8f5c3d6a0a828f7d6fbabcaf815415dd125c32533013f85603be46d'
             }

diff --git a/keras_ocr/tools.py b/keras_ocr/tools.py
@@ -455,7 +455,6 @@ def download_and_verify(url, sha256=None, cache_dir=None, verbose=True, filename
     return filepath
 
 
-# pylint: disable=bad-continuation
 def get_rotated_box(
     points
 ) -> typing.Tuple[typing.Tuple[float, float], typing.Tuple[float, float], typing.Tuple[

diff --git a/tests/test_pytorch_keras.py b/tests/test_pytorch_keras.py
@@ -17,11 +17,11 @@
 def test_pytorch_identical_output():
     import torch  # pylint: disable=import-outside-toplevel
     weights_path_torch = keras_ocr.tools.download_and_verify(
-        url='https://www.mediafire.com/file/qh2ullnnywi320s/craft_mlt_25k.pth/file',
+        url='https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/craft_mlt_25k.pth',
         filename='craft_mlt_25k.pth',
         sha256='4a5efbfb48b4081100544e75e1e2b57f8de3d84f213004b14b85fd4b3748db17')
     weights_path_keras = keras_ocr.tools.download_and_verify(
-        url='https://www.mediafire.com/file/mepzf3sq7u7nve9/craft_mlt_25k.h5/file',
+        url='https://github.com/faustomorales/keras-ocr/releases/download/v0.8.4/craft_mlt_25k.h5',
         filename='craft_mlt_25k.h5',
         sha256='7283ce2ff05a0617e9740c316175ff3bacdd7215dbdf1a726890d5099431f899')