Being nice to the linter

Author: andbue

calamari_ocr/ocr/dataset/datareader/abbyy/xml/data.py

@@ -1,11 +1,9 @@
 class Rect:
-
     """
     Class defines the rectangle of an element in the Abbyy document
     """
 
     def __init__(self, l: int, t: int, r: int, b: int):
-
         """
         Constructs a Rect class
         :param l: length
@@ -36,7 +34,6 @@ def __str__(self):
 
 
 class Book:
-
     """
     Main class; contains all subelements: book -> page -> block -> par -> line -> format
     """
@@ -64,7 +61,6 @@ def __str__(self):
                             s += "                  " + format.__str__()
 
     def getBlocks(self) -> []:
-
         """
         :return: All the blocks of this book
         """
@@ -78,7 +74,6 @@ def getBlocks(self) -> []:
         return blocks
 
     def getPars(self) -> []:
-
         """
         :return: All the paragraphs of this book
         """
@@ -93,7 +88,6 @@ def getPars(self) -> []:
         return pars
 
     def getLines(self) -> []:
-
         """
         :return: All the lines of this book
         """
@@ -109,7 +103,6 @@ def getLines(self) -> []:
         return lines
 
     def getFormats(self) -> []:
-
         """
         :return: All the chars of this book
         """
@@ -127,7 +120,6 @@ def getFormats(self) -> []:
 
 
 class Page:
-
     """
     Subelement of the book class; contains a list with the subelement block
     """
@@ -141,7 +133,6 @@ def __init__(
         imgFile: str,
         xmlFile: str,
     ):
-
         """
         Construct a page class with an empty block list
         :param width: The width of the page (in pixel)
@@ -178,7 +169,6 @@ def __str__(self):
         )
 
     def getPars(self) -> []:
-
         """
         :return: All the pars of this page
         """
@@ -192,7 +182,6 @@ def getPars(self) -> []:
         return pars
 
     def getLines(self) -> []:
-
         """
         :return: All the lines of this page
         """
@@ -207,7 +196,6 @@ def getLines(self) -> []:
         return lines
 
     def getFormats(self) -> []:
-
         """
         :return: All the Format Tags of this page
         """
@@ -224,13 +212,11 @@ def getFormats(self) -> []:
 
 
 class Block:
-
     """
     Subelement of the page class; contains a list with the subelement par
     """
 
     def __init__(self, blockType: str, blockName: str, rect: Rect):
-
         """
         Construct a block class with an empty line list
         :param blockType: The type of a block (further information in the abbyy doc)
@@ -246,7 +232,6 @@ def __str__(self):
         return "Block:[BlockType={}, rect={}]".format(self.blockType, self.rect)
 
     def getLines(self) -> []:
-
         """
         :return: All the lines of this block
         """
@@ -260,7 +245,6 @@ def getLines(self) -> []:
         return lines
 
     def getFormats(self) -> []:
-
         """
         :return: All the Format Tags of this block
         """
@@ -281,7 +265,6 @@ class Par:
     """
 
     def __init__(self, align: str, startIndent: int, lineSpacing: int):
-
         """
         Construct a Paragraph Class with an empty line list
         :param align:
@@ -306,7 +289,6 @@ def __str__(self):
         )
 
     def getFormats(self) -> []:
-
         """
         :return: All the Format Tags of the Paragraph
         """
@@ -321,13 +303,11 @@ def getFormats(self) -> []:
 
 
 class Line:
-
     """ "
     Subelement of the par class; contains a list with the subelement format
     """
 
     def __init__(self, baseline: int, rect: Rect):
-
         """
         Construct a line class with an empty char list
         :param baseline: ???

calamari_ocr/ocr/dataset/datareader/abbyy/xml/reader.py

@@ -15,7 +15,6 @@ class XMLReader:
     """
 
     def __init__(self, imgfiles: [], abbyfiles: [], skip_invalid: bool):
-
         """
         Constructs an XMLReader class with the :param directory
 
@@ -27,7 +26,6 @@ def __init__(self, imgfiles: [], abbyfiles: [], skip_invalid: bool):
         self.skip_invalid = skip_invalid
 
     def read(self) -> Book:
-
         """
         Start trying to read the data from the directory :var self.directory
 

calamari_ocr/ocr/dataset/datareader/abbyy/xml/writer.py

@@ -74,7 +74,6 @@ def write(page: Page, filename: str):
 
     @staticmethod
     def _addElement(element, key, value):
-
         """
         Only add attributes to an tag if the key is not None
 

calamari_ocr/ocr/model/ctcdecoder/token_passing_ctc_decoder.py

@@ -67,15 +67,15 @@ def get(self, w, s, t):
         return self.tokens[(w, s, t)]
 
     def dump(self, s, t):
-        for (k, v) in self.tokens.items():
+        for k, v in self.tokens.items():
             if k[1] == s and k[2] == t:
                 print(k, v)
 
 
 def outputIndices(toks, words, s, t):
     """argmax_w tok(w,s,t)"""
     res = []
-    for (wIdx, _) in enumerate(words):
+    for wIdx, _ in enumerate(words):
         res.append(toks.get(wIdx, s, t))
 
     idx = [i[0] for i in sorted(enumerate(res), key=lambda x: x[1].score)]
@@ -109,7 +109,7 @@ def ctcTokenPassing(mat, classes, charWords, blankIdx=-1, word_separator=" "):
     toks = TokenList()
 
     # Initialisation: 1-9
-    for (wIdx, w) in enumerate(words):
+    for wIdx, w in enumerate(words):
         w = words[wIdx]
         wPrime = primeWords[wIdx]
 

calamari_ocr/ocr/model/ensemblegraph.py

@@ -112,6 +112,7 @@ def gen_empty_output(bs):
                 }
 
             empty_output = gen_empty_output(1)
+
             # Validation: Compute output for each graph but only for its own partition
             # Per sample this is one CER which is then used e. g. for early stopping
             def apply_single_model(batch):

calamari_ocr/ocr/model/ensemblemodel.py

@@ -78,7 +78,12 @@ def cer(decoded, targets, targets_length):
         # using the true codec size (the W/B-Matrix may change its shape however during loading/codec change
         # to match the true codec size
         sw = K.flatten(targets["gt_len"])
-        return [self.cer_total(cer(outputs["decoded"], targets["gt"], targets["gt_len"]), sample_weight=sw,)] + [
+        return [
+            self.cer_total(
+                cer(outputs["decoded"], targets["gt"], targets["gt_len"]),
+                sample_weight=sw,
+            )
+        ] + [
             self.sub_cer[i](
                 cer(outputs[f"decoded_{i}"], targets["gt"], targets["gt_len"]),
                 sample_weight=sw * tf.cast(tf.equal(K.flatten(targets["fold_id"]), i), tf.int32),

calamari_ocr/scripts/experiment.py

@@ -17,6 +17,7 @@
 # path to the dir of this script to automatically detect the training script
 this_absdir = os.path.dirname(os.path.abspath(inspect.stack()[0][1]))
 
+
 # create necessary directories
 def run_for_single_line(args):
     # lines/network/pretraining as base dir

calamari_ocr/scripts/experiment_cross_fold.py

@@ -15,6 +15,7 @@
 # path to the dir of this script to automatically detect the training script
 this_absdir = os.path.dirname(os.path.abspath(inspect.stack()[0][1]))
 
+
 # create necessary directories
 def run_for_single_line(args):
     # lines/network/pretraining as base dir

calamari_ocr/thirdparty/ctcwordbeamsearch/Beam.py

@@ -163,14 +163,14 @@ def getBestBeams(self, num):
 
     def deletePartialBeams(self, lm):
         "delete beams for which last word is not finished"
-        for (k, v) in self.beams.items():
+        for k, v in self.beams.items():
             lastWord = v.textual.wordDev
             if (lastWord != "") and (not lm.isWord(lastWord)):
                 del self.beams[k]
 
     def completeBeams(self, lm):
         "complete beams such that last word is complete word"
-        for (_, v) in self.beams.items():
+        for _, v in self.beams.items():
             lastPrefix = v.textual.wordDev
             if lastPrefix == "" or lm.isWord(lastPrefix):
                 continue

calamari_ocr/thirdparty/ctcwordbeamsearch/main.py

@@ -28,7 +28,7 @@
     csv = Utils.CSVWriter()
 
     # decode each sample from dataset
-    for (idx, data) in enumerate(loader):
+    for idx, data in enumerate(loader):
         # decode matrix
         res = wordBeamSearch(data.mat, 10, loader.lm, useNGrams)
         print("Sample: " + str(idx + 1))