#471 variables renaming.

arekit/contrib/utils/pipelines/sources/ruattitudes/extract_text_opinions.py

@@ -39,20 +39,21 @@ def create_text_opinion_extraction_pipeline(text_parser,
     assert(version in [RuAttitudesVersions.V20Large, RuAttitudesVersions.V20Base,
                        RuAttitudesVersions.V20BaseNeut, RuAttitudesVersions.V20LargeNeut])
 
-    doc_ops = RuAttitudesDocumentProvider(version=version,
-                                          keep_doc_ids_only=False,
-                                          doc_id_func=lambda doc_id: doc_id,
-                                          limit=limit)
+    doc_provider = RuAttitudesDocumentProvider(version=version,
+                                               keep_doc_ids_only=False,
+                                               doc_id_func=lambda doc_id: doc_id,
+                                               limit=limit)
 
     pipeline = text_opinion_extraction_pipeline(
         annotators=[
-            PredefinedTextOpinionAnnotator(doc_ops=doc_ops, label_formatter=RuAttitudesLabelFormatter(label_scaler))
+            PredefinedTextOpinionAnnotator(doc_provider=doc_provider,
+                                           label_formatter=RuAttitudesLabelFormatter(label_scaler))
         ],
         text_opinion_filters=[
             EntityBasedTextOpinionFilter(entity_filter=entity_filter),
             DistanceLimitedTextOpinionFilter(terms_per_context)
         ],
-        get_doc_by_id_func=doc_ops.by_id,
+        get_doc_by_id_func=doc_provider.by_id,
         text_parser=text_parser)
 
     return pipeline

arekit/contrib/utils/pipelines/sources/rusentrel/extract_text_opinions.py

@@ -43,7 +43,7 @@ def create_text_opinion_extraction_pipeline(rusentrel_version,
         stemmer=MystemWrapper(),
         is_read_only=False)
 
-    doc_ops = RuSentrelDocumentProvider(version=rusentrel_version, synonyms=synonyms)
+    doc_provider = RuSentrelDocumentProvider(version=rusentrel_version, synonyms=synonyms)
 
     pipeline = text_opinion_extraction_pipeline(
         annotators=[
@@ -55,7 +55,7 @@ def create_text_opinion_extraction_pipeline(rusentrel_version,
             EntityBasedTextOpinionFilter(entity_filter=entity_filter),
             DistanceLimitedTextOpinionFilter(terms_per_context)
         ],
-        get_doc_by_id_func=doc_ops.by_id,
+        get_doc_by_id_func=doc_provider.by_id,
         text_parser=text_parser)
 
     return pipeline

arekit/contrib/utils/pipelines/sources/sentinerel/extract_text_opinions.py

@@ -25,7 +25,7 @@ def create_text_opinion_extraction_pipeline(sentinerel_version,
                                             text_parser,
                                             label_formatter=SentiNERELSentimentLabelFormatter(),
                                             terms_per_context=50,
-                                            doc_ops=None,
+                                            doc_provider=None,
                                             dist_in_sentences=0,
                                             docs_limit=None,
                                             entity_filter=None):
@@ -40,7 +40,7 @@ def create_text_opinion_extraction_pipeline(sentinerel_version,
             sentinerel_version: enum
                 Version of the SentiNEREl collection.
             text_parser: Is the way of how do we process the text.
-            doc_ops: DocumentProvider or None
+            doc_provider: DocumentProvider or None
                 In case of None we consider the default initialization.
             label_formatter:
                 Formatter for labels which allows to: limit set of labels, and perform its conversion from
@@ -52,16 +52,16 @@ def create_text_opinion_extraction_pipeline(sentinerel_version,
             pipelines per every type.
     """
     assert(isinstance(sentinerel_version, SentiNerelVersions))
-    assert(isinstance(doc_ops, DocumentProvider) or doc_ops is None)
+    assert(isinstance(doc_provider, DocumentProvider) or doc_provider is None)
 
     data_folding = None
 
-    if doc_ops is None:
+    if doc_provider is None:
         # Default Initialization.
         filenames_by_ids, data_folding = SentiNerelIOUtils.read_dataset_split(version=sentinerel_version,
                                                                               docs_limit=docs_limit)
-        doc_ops = SentiNERELDocProvider(filename_by_id=filenames_by_ids,
-                                        version=sentinerel_version)
+        doc_provider = SentiNERELDocProvider(filename_by_id=filenames_by_ids,
+                                             version=sentinerel_version)
 
     train_neut_annot = create_nolabel_text_opinion_annotator(terms_per_context=terms_per_context,
                                                              dist_in_sents=dist_in_sentences)
@@ -74,28 +74,28 @@ def create_text_opinion_extraction_pipeline(sentinerel_version,
         DistanceLimitedTextOpinionFilter(terms_per_context)
     ]
 
-    predefined_annot = PredefinedTextOpinionAnnotator(doc_ops, label_formatter)
+    predefined_annot = PredefinedTextOpinionAnnotator(doc_provider, label_formatter)
 
     pipelines = {
         DataType.Train: create_train_pipeline(text_parser=text_parser,
-                                              doc_ops=doc_ops,
+                                              doc_provider=doc_provider,
                                               annotators=[
                                                   predefined_annot,
                                                   train_neut_annot
                                               ],
                                               text_opinion_filters=text_opinion_filters),
         DataType.Test: create_test_pipeline(text_parser=text_parser,
-                                            doc_ops=doc_ops,
+                                            doc_provider=doc_provider,
                                             annotators=[
                                                 test_neut_annot
                                             ],
                                             text_opinion_filters=text_opinion_filters),
         DataType.Etalon: create_etalon_pipeline(text_parser=text_parser,
-                                                doc_ops=doc_ops,
+                                                doc_provider=doc_provider,
                                                 predefined_annot=predefined_annot,
                                                 text_opinion_filters=text_opinion_filters),
         DataType.Dev: create_etalon_with_no_label_pipeline(text_parser=text_parser,
-                                                           doc_ops=doc_ops,
+                                                           doc_provider=doc_provider,
                                                            annotators=[
                                                                predefined_annot,
                                                                train_neut_annot
@@ -142,47 +142,47 @@ def create_nolabel_text_opinion_annotator(terms_per_context, dist_in_sents=0, sy
             error_on_synonym_end_missed=False))
 
 
-def create_train_pipeline(text_parser, doc_ops, annotators, text_opinion_filters):
+def create_train_pipeline(text_parser, doc_provider, annotators, text_opinion_filters):
     """ Train pipeline is based on the predefined annotations and
         automatic annotations of other pairs with a NoLabel.
     """
     return text_opinion_extraction_pipeline(
-        get_doc_by_id_func=doc_ops.by_id,
+        get_doc_by_id_func=doc_provider.by_id,
         text_parser=text_parser,
         annotators=annotators,
         text_opinion_filters=text_opinion_filters)
 
 
-def create_test_pipeline(text_parser, doc_ops, annotators, text_opinion_filters):
+def create_test_pipeline(text_parser, doc_provider, annotators, text_opinion_filters):
     """ This is a pipeline for TEST data annotation.
         We perform annotation of the attitudes.
     """
     assert(isinstance(text_parser, BaseTextParser))
     assert(isinstance(annotators, list))
-    assert(isinstance(doc_ops, DocumentProvider))
+    assert(isinstance(doc_provider, DocumentProvider))
 
     return text_opinion_extraction_pipeline(
         annotators=annotators,
         text_parser=text_parser,
-        get_doc_by_id_func=doc_ops.by_id,
+        get_doc_by_id_func=doc_provider.by_id,
         text_opinion_filters=text_opinion_filters)
 
 
-def create_etalon_pipeline(text_parser, doc_ops, predefined_annot, text_opinion_filters):
+def create_etalon_pipeline(text_parser, doc_provider, predefined_annot, text_opinion_filters):
     """ We adopt exact the same pipeline as for training data,
         but we do not perform "NoLabel" annotation.
         (we are interested only in sentiment attitudes).
     """
     return create_train_pipeline(text_parser=text_parser,
-                                 doc_ops=doc_ops,
+                                 doc_provider=doc_provider,
                                  annotators=[predefined_annot],
                                  text_opinion_filters=text_opinion_filters)
 
 
-def create_etalon_with_no_label_pipeline(annotators, text_parser, doc_ops, text_opinion_filters):
+def create_etalon_with_no_label_pipeline(annotators, text_parser, doc_provider, text_opinion_filters):
     """ We adopt exact the same pipeline as for training data.
     """
     return create_train_pipeline(text_parser=text_parser,
-                                 doc_ops=doc_ops,
+                                 doc_provider=doc_provider,
                                  annotators=annotators,
                                  text_opinion_filters=text_opinion_filters)

arekit/contrib/utils/pipelines/text_opinion/annot/predefined.py

@@ -14,7 +14,7 @@ class PredefinedTextOpinionAnnotator(BaseOpinionAnnotator):
         It converts the pre-annotated Relations from BRAT-documents to TextOpinions
     """
 
-    def __init__(self, doc_ops, label_formatter, keep_any_type=False, entity_index_func=None):
+    def __init__(self, doc_provider, label_formatter, keep_any_type=False, entity_index_func=None):
         """
             get_doc_func:
                 func(doc_id)
@@ -29,12 +29,12 @@ def __init__(self, doc_ops, label_formatter, keep_any_type=False, entity_index_f
             entity_index_func: is a way of how we provide an external entity ID
                 fund(entity) -> ID
         """
-        assert(isinstance(doc_ops, DocumentProvider))
+        assert(isinstance(doc_provider, DocumentProvider))
         assert(isinstance(label_formatter, StringLabelsFormatter))
         assert(callable(entity_index_func) or entity_index_func is None)
         super(PredefinedTextOpinionAnnotator, self).__init__()
 
-        self.__doc_ops = doc_ops
+        self.__doc_provider = doc_provider
         self.__label_formatter = label_formatter
         self.__keep_any_type = keep_any_type
         self.__entity_index_func = (lambda brat_entity: brat_entity.ID) if \
@@ -66,7 +66,7 @@ def _annot_collection_core(self, parsed_doc):
             EntityServiceProvider(self.__entity_index_func)
         ])
         esp = pns.get_provider(EntityServiceProvider.NAME)
-        doc = self.__doc_ops.by_id(parsed_doc.RelatedDocID)
+        doc = self.__doc_provider.by_id(parsed_doc.RelatedDocID)
 
         for brat_relation in doc.Relations:
 

tests/contrib/utils/test_csv_stream_write.py

@@ -66,19 +66,19 @@ def __launch(self, writer, target_extention):
         # Declaring pipeline related context parameters.
         #####
         no_folding = NoFolding(doc_ids=[0, 1], supported_data_type=DataType.Train)
-        doc_ops = FooDocumentProvider()
+        doc_provider = FooDocumentProvider()
         text_parser = BaseTextParser(pipeline=[BratTextEntitiesParser(), DefaultTextTokenizer(keep_tokens=True)])
         train_pipeline = text_opinion_extraction_pipeline(
             annotators=[
                 PredefinedTextOpinionAnnotator(
-                    doc_ops,
+                    doc_provider,
                     label_formatter=CustomLabelsFormatter(pos_label_type=Positive,
                                                           neg_label_type=Negative))
             ],
             text_opinion_filters=[
                 DistanceLimitedTextOpinionFilter(terms_per_context=50)
             ],
-            get_doc_by_id_func=doc_ops.by_id,
+            get_doc_by_id_func=doc_provider.by_id,
             text_parser=text_parser)
         #####
 

tests/tutorials/test_tutorial_data_foldings.py

@@ -41,11 +41,11 @@ def test(self):
 
         splitter_simple = SimpleCrossValidationSplitter(shuffle=True, seed=1)
 
-        doc_ops = FooDocumentProvider()
+        doc_provider = FooDocumentProvider()
         doc_ids = list(range(2))
 
         splitter_statistical = StatBasedCrossValidationSplitter(
-            docs_stat=SentenceBasedDocumentStatGenerator(doc_reader_func=doc_ops.by_id),
+            docs_stat=SentenceBasedDocumentStatGenerator(doc_reader_func=doc_provider.by_id),
             doc_ids=doc_ids)
 
         cv_folding = TwoClassCVFolding(

tests/tutorials/test_tutorial_pipeline_sampling_bert.py

@@ -105,19 +105,19 @@ def test(self):
         # Declaring pipeline related context parameters.
         #####
         no_folding = NoFolding(doc_ids=[0, 1], supported_data_type=DataType.Train)
-        doc_ops = FooDocumentProvider()
+        doc_provider = FooDocumentProvider()
         text_parser = BaseTextParser(pipeline=[BratTextEntitiesParser(), DefaultTextTokenizer(keep_tokens=True)])
         train_pipeline = text_opinion_extraction_pipeline(
             annotators=[
                 PredefinedTextOpinionAnnotator(
-                    doc_ops,
+                    doc_provider,
                     label_formatter=CustomLabelsFormatter(pos_label_type=Positive,
                                                           neg_label_type=Negative))
             ],
             text_opinion_filters=[
                 DistanceLimitedTextOpinionFilter(terms_per_context=50)
             ],
-            get_doc_by_id_func=doc_ops.by_id,
+            get_doc_by_id_func=doc_provider.by_id,
             text_parser=text_parser)
         #####
 

tests/tutorials/test_tutorial_pipeline_sampling_network.py

@@ -106,7 +106,7 @@ def test(self):
         # Declaring pipeline related context parameters.
         #####
         no_folding = NoFolding(doc_ids=[0, 1], supported_data_type=DataType.Train)
-        doc_ops = FooDocumentProvider()
+        doc_provider = FooDocumentProvider()
         text_parser = BaseTextParser(pipeline=[
             BratTextEntitiesParser(),
             DefaultTextTokenizer(keep_tokens=True),
@@ -115,13 +115,13 @@ def test(self):
         train_pipeline = text_opinion_extraction_pipeline(
             annotators=[
                 PredefinedTextOpinionAnnotator(
-                    doc_ops,
+                    doc_provider,
                     label_formatter=CustomLabelsFormatter(pos_label_type=Positive, neg_label_type=Negative))
             ],
             text_opinion_filters=[
                 DistanceLimitedTextOpinionFilter(terms_per_context=50)
             ],
-            get_doc_by_id_func=doc_ops.by_id,
+            get_doc_by_id_func=doc_provider.by_id,
             text_parser=text_parser)
         #####
 

tests/tutorials/test_tutorial_pipeline_text_opinion_annotation.py

@@ -44,9 +44,9 @@ def __init__(self, pos_label_type, neg_label_type):
 class TestTextOpinionAnnotation(unittest.TestCase):
 
     def test(self):
-        doc_ops = FooDocumentProvider()
+        doc_provider = FooDocumentProvider()
         predefined_annotator = PredefinedTextOpinionAnnotator(
-            doc_ops=doc_ops,
+            doc_provider=doc_provider,
             label_formatter=CustomLabelsFormatter(pos_label_type=PositiveLabel,
                                                   neg_label_type=NegativeLabel))
 
@@ -76,7 +76,7 @@ def test(self):
             text_opinion_filters=[
                 DistanceLimitedTextOpinionFilter(terms_per_context=50)
             ],
-            get_doc_by_id_func=doc_ops.by_id,
+            get_doc_by_id_func=doc_provider.by_id,
             text_parser=text_parser)
 
         # Running the pipeline.