Update files from the datasets library (from 1.6.0)

Release notes: https://github.com/huggingface/datasets/releases/tag/1.6.0

README.md

@@ -429,7 +429,10 @@ WikiANN (sometimes called PAN-X) is a multilingual named entity recognition data
 
 ### Data Fields
 
-[More Information Needed]
+- `tokens`: a `list` of `string` features.
+- `langs`: a `list` of `string` features that correspond to the language of each token.
+- `ner_tags`: a `list` of classification labels, with possible values including `O` (0), `B-PER` (1), `I-PER` (2), `B-ORG` (3), `I-ORG` (4), `B-LOC` (5), `I-LOC` (6).
+- `spans`: a `list` of `string` features, that is the list of named entities in the input text formatted as ``<TAG>: <mention>``
 
 ### Data Splits
 
@@ -535,4 +538,4 @@ while the 176 languages supported in this version are associated with the follow
 
 ### Contributions
 
-Thanks to [@lewtun](https://github.com/lewtun) for adding this dataset.
+Thanks to [@lewtun](https://github.com/lewtun) and [@rabeehk](https://github.com/rabeehk) for adding this dataset.

wikiann.py

@@ -15,7 +15,6 @@
 
 """The WikiANN dataset for multilingual named entity recognition"""
 
-from __future__ import absolute_import, division, print_function
 
 import os
 
@@ -241,6 +240,61 @@ class Wikiann(datasets.GeneratorBasedBuilder):
         WikiannConfig(name=lang, description=f"WikiANN NER examples in language {lang}") for lang in _LANGS
     ]
 
+    def _tags_to_spans(self, tags):
+        """Convert tags to spans."""
+        spans = set()
+        span_start = 0
+        span_end = 0
+        active_conll_tag = None
+        for index, string_tag in enumerate(tags):
+            # Actual BIO tag.
+            bio_tag = string_tag[0]
+            assert bio_tag in ["B", "I", "O"], "Invalid Tag"
+            conll_tag = string_tag[2:]
+            if bio_tag == "O":
+                # The span has ended.
+                if active_conll_tag:
+                    spans.add((active_conll_tag, (span_start, span_end)))
+                active_conll_tag = None
+                # We don't care about tags we are
+                # told to ignore, so we do nothing.
+                continue
+            elif bio_tag == "B":
+                # We are entering a new span; reset indices and active tag to new span.
+                if active_conll_tag:
+                    spans.add((active_conll_tag, (span_start, span_end)))
+                active_conll_tag = conll_tag
+                span_start = index
+                span_end = index
+            elif bio_tag == "I" and conll_tag == active_conll_tag:
+                # We're inside a span.
+                span_end += 1
+            else:
+                # This is the case the bio label is an "I", but either:
+                # 1) the span hasn't started - i.e. an ill formed span.
+                # 2) We have IOB1 tagging scheme.
+                # We'll process the previous span if it exists, but also include this
+                # span. This is important, because otherwise, a model may get a perfect
+                # F1 score whilst still including false positive ill-formed spans.
+                if active_conll_tag:
+                    spans.add((active_conll_tag, (span_start, span_end)))
+                active_conll_tag = conll_tag
+                span_start = index
+                span_end = index
+        # Last token might have been a part of a valid span.
+        if active_conll_tag:
+            spans.add((active_conll_tag, (span_start, span_end)))
+        # Return sorted list of spans
+        return sorted(list(spans), key=lambda x: x[1][0])
+
+    def _get_spans(self, tokens, tags):
+        """Convert tags to textspans."""
+        spans = self._tags_to_spans(tags)
+        text_spans = [x[0] + ": " + " ".join([tokens[i] for i in range(x[1][0], x[1][1] + 1)]) for x in spans]
+        if not text_spans:
+            text_spans = ["None"]
+        return text_spans
+
     def _info(self):
         features = datasets.Features(
             {
@@ -259,6 +313,7 @@ class Wikiann(datasets.GeneratorBasedBuilder):
                     )
                 ),
                 "langs": datasets.Sequence(datasets.Value("string")),
+                "spans": datasets.Sequence(datasets.Value("string")),
             }
         )
         return datasets.DatasetInfo(
@@ -290,6 +345,26 @@ class Wikiann(datasets.GeneratorBasedBuilder):
         ]
 
     def _generate_examples(self, filepath):
+        """Reads line by line format of the NER dataset and generates examples.
+        Input Format:
+        en:rick  B-PER
+        en:and  O
+        en:morty  B-PER
+        en:are  O
+        en:cool  O
+        en:.  O
+        Output Format:
+        {
+        'tokens': ["rick", "and", "morty", "are", "cool", "."],
+        'ner_tags': ["B-PER", "O" , "B-PER", "O", "O", "O"],
+        'langs': ["en", "en", "en", "en", "en", "en"]
+        'spans': ["PER: rick", "PER: morty"]
+        }
+        Args:
+            filepath: Path to file with line by line NER format.
+        Returns:
+            Examples with the format listed above.
+        """
         guid_index = 1
         with open(filepath, encoding="utf-8") as f:
             tokens = []
@@ -298,7 +373,8 @@ class Wikiann(datasets.GeneratorBasedBuilder):
             for line in f:
                 if line == "" or line == "\n":
                     if tokens:
-                        yield guid_index, {"tokens": tokens, "ner_tags": ner_tags, "langs": langs}
+                        spans = self._get_spans(tokens, ner_tags)
+                        yield guid_index, {"tokens": tokens, "ner_tags": ner_tags, "langs": langs, "spans": spans}
                         guid_index += 1
                         tokens = []
                         ner_tags = []