Merge pull request #567 from stweil/PEP8

Clean import statements

docs/conf.py

@@ -12,11 +12,8 @@
 # All configuration values have a default; values that are commented out
 # serve to show the default.
 
-import sys
-import os
-import shlex
+from subprocess import PIPE, Popen
 
-from subprocess import Popen, PIPE
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.

kraken/align.py

@@ -21,18 +21,17 @@
 A character alignment module using a network output lattice and ground truth to
 accuractely determine grapheme locations in input data.
 """
-import torch
-import logging
 import dataclasses
+import logging
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Literal, Optional
 
-from PIL import Image
+import torch
 from bidi.algorithm import get_display
-
-from dataclasses import dataclass
-from typing import Optional, Literal, TYPE_CHECKING
+from PIL import Image
 
 from kraken import rpred
-from kraken.containers import Segmentation, BaselineOCRRecord
+from kraken.containers import BaselineOCRRecord, Segmentation
 
 if TYPE_CHECKING:
     from kraken.lib.models import TorchSeqRecognizer

kraken/binarization.py

@@ -19,17 +19,17 @@
 
 An adaptive binarization algorithm.
 """
-import warnings
 import logging
-import numpy as np
+import warnings
+from typing import TYPE_CHECKING
 
-from kraken.lib.util import pil2array, array2pil, is_bitonal, get_im_str
-from scipy.ndimage import affine_transform, percentile_filter, gaussian_filter, binary_dilation
+import numpy as np
+from scipy.ndimage import (affine_transform, binary_dilation, gaussian_filter,
+                           percentile_filter)
 from scipy.ndimage import zoom as _zoom
 
-from typing import TYPE_CHECKING
-
 from kraken.lib.exceptions import KrakenInputException
+from kraken.lib.util import array2pil, get_im_str, is_bitonal, pil2array
 
 if TYPE_CHECKING:
     from PIL import Image

kraken/blla.py

@@ -20,32 +20,30 @@
 line recognizer uses the baseline paradigm.
 """
 
-import PIL
-import uuid
-import torch
 import logging
-import numpy as np
+import uuid
+from typing import Any, Callable, Dict, List, Literal, Optional, Union
+
 import importlib_resources
+import numpy as np
+import PIL
 import shapely.geometry as geom
+import torch
 import torch.nn.functional as F
 import torchvision.transforms as tf
-
-from typing import Optional, Dict, Callable, Union, List, Any, Literal
-
 from scipy.ndimage import gaussian_filter
 from skimage.filters import sobel
 
-from kraken.lib import vgsl, dataset
-from kraken.containers import Region, Segmentation, BaselineLine
-from kraken.lib.util import is_bitonal, get_im_str
-from kraken.lib.exceptions import KrakenInputException, KrakenInvalidModelException
-from kraken.lib.segmentation import (polygonal_reading_order,
-                                     neural_reading_order,
-                                     vectorize_lines, vectorize_regions,
-                                     scale_polygonal_lines,
-                                     calculate_polygonal_environment,
-                                     is_in_region,
-                                     scale_regions)
+from kraken.containers import BaselineLine, Region, Segmentation
+from kraken.lib import dataset, vgsl
+from kraken.lib.exceptions import (KrakenInputException,
+                                   KrakenInvalidModelException)
+from kraken.lib.segmentation import (calculate_polygonal_environment,
+                                     is_in_region, neural_reading_order,
+                                     polygonal_reading_order,
+                                     scale_polygonal_lines, scale_regions,
+                                     vectorize_lines, vectorize_regions)
+from kraken.lib.util import get_im_str, is_bitonal
 
 __all__ = ['segment']
 

kraken/containers.py

@@ -19,12 +19,13 @@
 Container classes replacing the old dictionaries returned by kraken's
 functional blocks.
 """
-import numpy as np
-import bidi.algorithm as bd
-
-from typing import Literal, List, Dict, Union, Optional, Tuple, Any, TYPE_CHECKING
-from dataclasses import dataclass, asdict
 from abc import ABC, abstractmethod
+from dataclasses import asdict, dataclass
+from typing import (TYPE_CHECKING, Any, Dict, List, Literal, Optional, Tuple,
+                    Union)
+
+import bidi.algorithm as bd
+import numpy as np
 
 from kraken.lib.segmentation import compute_polygon_section
 

kraken/contrib/baselineset_overlay.py

@@ -9,10 +9,12 @@
 @click.argument('files', nargs=-1)
 def cli(files):
 
-    import torch
-    from PIL import Image
     from os.path import splitext
+
+    import torch
     import torchvision.transforms as tf
+    from PIL import Image
+
     from kraken.lib import dataset
 
     batch, channels, height, width = 1, 3, 1200, 0

kraken/contrib/extract_lines.py

@@ -20,13 +20,15 @@ def cli(format_type, model, files):
         click.echo(ctx.get_help())
         ctx.exit()
 
-    from PIL import Image
-    from os.path import splitext
-    from kraken import blla
-    from kraken.lib import segmentation, vgsl, xml
     import io
     import json
+    from os.path import splitext
+
     import pyarrow as pa
+    from PIL import Image
+
+    from kraken import blla
+    from kraken.lib import segmentation, vgsl, xml
 
     if model is None:
         for doc in files:

kraken/contrib/forced_alignment_overlay.py

@@ -3,14 +3,15 @@
 Draws a transparent overlay of the forced alignment output over the input
 image.
 """
-import re
 import os
-import click
+import re
 import unicodedata
-from lxml import etree
 from itertools import cycle
 from unicodedata import normalize
 
+import click
+from lxml import etree
+
 cmap = cycle([(230, 25, 75, 127),
               (60, 180, 75, 127),
               (255, 225, 25, 127),
@@ -100,9 +101,9 @@ def cli(format_type, model, normalization, output, files):
 
     from PIL import Image, ImageDraw
 
-    from kraken.lib.xml import XMLPage
-    from kraken.lib import models
     from kraken import align
+    from kraken.lib import models
+    from kraken.lib.xml import XMLPage
 
     if format_type == 'alto':
         repl_fn = _repl_alto

kraken/contrib/generate_scripts.py

@@ -2,8 +2,9 @@
 """
 Script fetching the latest unicode Scripts.txt and dumping it as json.
 """
-from urllib import request
 import json
+from urllib import request
+
 import regex
 
 uri = 'http://www.unicode.org/Public/UNIDATA/Scripts.txt'

kraken/contrib/heatmap_overlay.py

@@ -14,12 +14,14 @@ def cli(model, files):
     Applies a BLLA baseline segmentation model and outputs the raw heatmaps of
     the first baseline class.
     """
+    from os.path import splitext
+
     import torch
-    from PIL import Image
-    from kraken.lib import vgsl, dataset
     import torch.nn.functional as F
-    from os.path import splitext
     import torchvision.transforms as tf
+    from PIL import Image
+
+    from kraken.lib import dataset, vgsl
 
     model = vgsl.TorchVGSLModel.load_model(model)
     model.eval()

kraken/contrib/hyperparameters/tune_pretraining.py

@@ -2,19 +2,19 @@
 """
 A script for a grid search over pretraining hyperparameters.
 """
-import click
 from functools import partial
 
+import click
+import pytorch_lightning as pl
+from pytorch_lightning import seed_everything
 from ray import tune
-
 from ray.tune.integration.pytorch_lightning import TuneReportCallback
 
-from kraken.lib.default_specs import RECOGNITION_PRETRAIN_HYPER_PARAMS, RECOGNITION_SPEC
-from kraken.lib.pretrain.model import PretrainDataModule, RecognitionPretrainModel
 from kraken.ketos.util import _validate_manifests
-
-import pytorch_lightning as pl
-from pytorch_lightning import seed_everything
+from kraken.lib.default_specs import (RECOGNITION_PRETRAIN_HYPER_PARAMS,
+                                      RECOGNITION_SPEC)
+from kraken.lib.pretrain.model import (PretrainDataModule,
+                                       RecognitionPretrainModel)
 
 config = {'lrate': tune.loguniform(1e-8, 1e-2),
           'num_negatives': tune.qrandint(1, 4, 1),

kraken/contrib/hyperparameters/tune_training.py

@@ -5,15 +5,14 @@
 import sys
 from functools import partial
 
+import pytorch_lightning as pl
 from ray import tune
-
 from ray.tune.integration.pytorch_lightning import TuneReportCallback
 
-from kraken.lib.default_spec import RECOGNITION_PRETRAIN_HYPER_PARAMS, RECOGNITION_SPEC
-from kraken.lib.pretrain.model import PretrainDataModule, RecognitionPretrainModel
-
-import pytorch_lightning as pl
-
+from kraken.lib.default_spec import (RECOGNITION_PRETRAIN_HYPER_PARAMS,
+                                     RECOGNITION_SPEC)
+from kraken.lib.pretrain.model import (PretrainDataModule,
+                                       RecognitionPretrainModel)
 
 config = {'lrate': tune.loguniform(1e-8, 1e-2),
           'num_negatives': tune.qrandint(2, 100, 8),

kraken/contrib/print_word_spreader.py

@@ -1,11 +1,16 @@
 #!/usr/bin/env python
 #2020, Bruce Robertson
 #Master file at https://github.com/brobertson/Lace2-tools/blob/master/normalize_hocr.py
-import html, os, sys, argparse
+import argparse
+import html
+import os
+import sys
 from statistics import mean
+
 from lxml import etree
 from PIL import Image
 
+
 #a custom exception to indicate when a page or other element doesn't
 #have a bounding box where we would expect it.
 class BboxError(Exception):

kraken/contrib/recognition_boxes.py

@@ -6,14 +6,14 @@
 
 import os
 import sys
+from itertools import cycle
 
 from PIL import Image, ImageDraw
 
-from kraken.pageseg import segment
 from kraken.binarization import nlbin
-from kraken.rpred import rpred
-from itertools import cycle
 from kraken.lib import models
+from kraken.pageseg import segment
+from kraken.rpred import rpred
 
 cmap = cycle([(230, 25, 75, 127),
               (60, 180, 75, 127),

kraken/contrib/repolygonize.py

@@ -28,12 +28,13 @@ def cli(format_type, topline, files):
         click.echo(ctx.get_help())
         ctx.exit()
 
-    from lxml import etree
+    from itertools import groupby
     from os.path import splitext
 
-    from itertools import groupby
-    from kraken.lib import xml
+    from lxml import etree
     from PIL import Image
+
+    from kraken.lib import xml
     from kraken.lib.segmentation import calculate_polygonal_environment
 
     def _repl_alto(fname, polygons):

kraken/contrib/segmentation_overlay.py

@@ -3,14 +3,14 @@
 Draws a transparent overlay of baseline segmenter output over a list of image
 files.
 """
-import re
+import dataclasses
 import os
-import click
+import re
 import unicodedata
-import dataclasses
-from itertools import cycle
 from collections import defaultdict
+from itertools import cycle
 
+import click
 
 cmap = cycle([(230, 25, 75, 127),
               (60, 180, 75, 127)])
@@ -58,8 +58,8 @@ def cli(model, text_direction, repolygonize, files):
 
     from PIL import Image, ImageDraw
 
-    from kraken.lib import vgsl, xml, segmentation
     from kraken import blla
+    from kraken.lib import segmentation, vgsl, xml
 
     if model is None:
         for doc in files:

kraken/contrib/set_seg_options.py

@@ -2,9 +2,10 @@
 """
 A script setting the metadata of segmentation models.
 """
-import click
 import shutil
 
+import click
+
 
 @click.command()
 @click.option('-b', '--bounding-region', multiple=True, help='Sets region identifiers which bound line bounding polygons')

kraken/ketos/__init__.py

@@ -19,9 +19,9 @@
 Command line drivers for training functionality.
 """
 
-import click
 import logging
 
+import click
 from PIL import Image
 from rich.traceback import install
 
@@ -30,11 +30,11 @@
 from .dataset import compile
 from .linegen import line_generator
 from .pretrain import pretrain
-from .recognition import train, test
+from .recognition import test, train
 from .repo import publish
-from .segmentation import segtrain, segtest
+from .ro import roadd, rotrain
+from .segmentation import segtest, segtrain
 from .transcription import extract, transcription
-from .ro import rotrain, roadd
 
 APP_NAME = 'kraken'