Merge pull request #2718 from flairNLP/GH-2717-ignore-labels

GH-2717: Add option to ignore labels in dataset

Author: alanakbik

flair/datasets/relation_extraction.py

@@ -39,6 +39,7 @@ def __init__(
         base_path: Union[str, Path] = None,
         in_memory: bool = True,
         augment_train: bool = False,
+        **corpusargs,
     ):
         """
         SemEval-2010 Task 8 on Multi-Way Classification of Semantic Relations Between Pairs of
@@ -83,6 +84,7 @@ def __init__(
             column_format={1: "text", 2: "ner"},
             comment_symbol="# ",
             in_memory=in_memory,
+            **corpusargs,
         )
 
     def extract_and_convert_to_conllu(self, data_file, data_folder, augment_train):
@@ -227,7 +229,7 @@ def _semeval_lines_to_token_list(self, raw_lines, augment_relations):
 
 
 class RE_ENGLISH_TACRED(ColumnCorpus):
-    def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True):
+    def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, **corpusargs):
         """
         TAC Relation Extraction Dataset with 41 relations from https://nlp.stanford.edu/projects/tacred/.
         Manual download is required for this dataset.
@@ -260,6 +262,7 @@ def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True):
             column_format={1: "text", 2: "ner"},
             comment_symbol="# ",
             in_memory=in_memory,
+            **corpusargs,
         )
 
     def extract_and_convert_to_conllu(self, data_file, data_folder):
@@ -351,7 +354,7 @@ def _tacred_example_to_token_list(self, example: Dict[str, Any]) -> conllu.Token
 
 
 class RE_ENGLISH_CONLL04(ColumnCorpus):
-    def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True):
+    def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True, **corpusargs):
         if not base_path:
             base_path = flair.cache_root / "datasets"
         else:
@@ -385,6 +388,7 @@ def __init__(self, base_path: Union[str, Path] = None, in_memory: bool = True):
             in_memory=in_memory,
             column_format={1: "text", 2: "ner"},
             comment_symbol="# ",
+            **corpusargs,
         )
 
     def _parse_incr(self, source_file) -> Iterable[conllu.TokenList]:
@@ -536,6 +540,7 @@ def __init__(
         base_path: Union[str, Path] = None,
         in_memory: bool = True,
         sentence_splitter: SentenceSplitter = SegtokSentenceSplitter(),
+        **corpusargs,
     ):
         """
         DrugProt corpus: Biocreative VII Track 1 from https://zenodo.org/record/5119892#.YSdSaVuxU5k/ on
@@ -570,6 +575,7 @@ def __init__(
             sample_missing_splits=False,
             column_format={1: "text", 2: "ner", 3: "ner"},
             comment_symbol="# ",
+            **corpusargs,
         )
 
     def extract_and_convert_to_conllu(self, data_file, data_folder):

flair/datasets/sequence_labeling.py

@@ -662,7 +662,8 @@ def _convert_lines_to_sentence(
                     for span_indices, score, label in predicted_spans:
                         span = sentence[span_indices[0] : span_indices[-1] + 1]
                         value = self._remap_label(label)
-                        span.add_label(span_level_tag_columns[span_column], value=value, score=score)
+                        if value != "O":
+                            span.add_label(span_level_tag_columns[span_column], value=value, score=score)
                 except Exception:
                     pass
 
@@ -681,7 +682,9 @@ def _convert_lines_to_sentence(
                     relation = Relation(
                         first=sentence[head_start - 1 : head_end], second=sentence[tail_start - 1 : tail_end]
                     )
-                    relation.add_label(typename="relation", value=self._remap_label(label))
+                    remapped = self._remap_label(label)
+                    if remapped != "O":
+                        relation.add_label(typename="relation", value=remapped)
 
         if len(sentence) > 0:
             return sentence
@@ -719,15 +722,17 @@ def _parse_token(self, line: str, column_name_map: Dict[int, str], last_token: O
                                 # add each other feature as label-value pair
                                 label_name = feature.split("=")[0]
                                 label_value = self._remap_label(feature.split("=")[1])
-                                token.add_label(label_name, label_value)
+                                if label_value != "O":
+                                    token.add_label(label_name, label_value)
 
                     else:
                         # get the task name (e.g. 'ner')
                         label_name = column_name_map[column]
                         # get the label value
                         label_value = self._remap_label(fields[column])
                         # add label
-                        token.add_label(label_name, label_value)
+                        if label_value != "O":
+                            token.add_label(label_name, label_value)
 
                 if column_name_map[column] == self.SPACE_AFTER_KEY and fields[column] == "-":
                     token.whitespace_after = False