Convert BIO tags to HuggingFace classes.

MASSIVE.py

@@ -260,6 +260,8 @@ _SCENARIOS = ['social', 'transport', 'calendar', 'play', 'news', 'datetime', 're
 
 _INTENTS = ['datetime_query', 'iot_hue_lightchange', 'transport_ticket', 'takeaway_query', 'qa_stock', 'general_greet', 'recommendation_events', 'music_dislikeness', 'iot_wemo_off', 'cooking_recipe', 'qa_currency', 'transport_traffic', 'general_quirky', 'weather_query', 'audio_volume_up', 'email_addcontact', 'takeaway_order', 'email_querycontact', 'iot_hue_lightup', 'recommendation_locations', 'play_audiobook', 'lists_createoradd', 'news_query', 'alarm_query', 'iot_wemo_on', 'general_joke', 'qa_definition', 'social_query', 'music_settings', 'audio_volume_other', 'calendar_remove', 'iot_hue_lightdim', 'calendar_query', 'email_sendemail', 'iot_cleaning', 'audio_volume_down', 'play_radio', 'cooking_query', 'datetime_convert', 'qa_maths', 'iot_hue_lightoff', 'iot_hue_lighton', 'transport_query', 'music_likeness', 'email_query', 'play_music', 'audio_volume_mute', 'social_post', 'alarm_set', 'qa_factoid', 'calendar_set', 'play_game', 'alarm_remove', 'lists_remove', 'transport_taxi', 'recommendation_movies', 'iot_coffee', 'music_query', 'play_podcasts', 'lists_query']
 
+_TAGS = ['I-playlist_name', 'I-general_frequency', 'B-audiobook_name', 'I-player_setting', 'I-business_type', 'I-time', 'I-place_name', 'I-date', 'B-device_type', 'I-song_name', 'B-timeofday', 'B-movie_type', 'B-order_type', 'B-date', 'B-news_topic', 'I-music_descriptor', 'B-media_type', 'B-cooking_type', 'B-meal_type', 'I-movie_name', 'B-joke_type', 'I-media_type', 'B-list_name', 'I-podcast_descriptor', 'I-meal_type', 'I-transport_descriptor', 'I-transport_agency', 'B-player_setting', 'B-house_place', 'B-music_genre', 'I-timeofday', 'I-personal_info', 'I-definition_word', 'B-podcast_name', 'I-podcast_name', 'I-music_album', 'I-transport_type', 'B-business_name', 'B-transport_name', 'B-sport_type', 'I-house_place', 'I-movie_type', 'B-transport_descriptor', 'B-artist_name', 'O', 'I-email_folder', 'I-event_name', 'B-email_folder', 'I-cooking_type', 'B-music_album', 'I-coffee_type', 'I-alarm_type', 'B-game_type', 'I-audiobook_name', 'B-playlist_name', 'B-alarm_type', 'B-place_name', 'I-relation', 'B-drink_type', 'I-drink_type', 'I-business_name', 'I-artist_name', 'B-music_descriptor', 'B-change_amount', 'I-weather_descriptor', 'I-game_name', 'I-app_name', 'I-ingredient', 'B-song_name', 'B-weather_descriptor', 'I-email_address', 'B-time', 'B-color_type', 'B-food_type', 'I-person', 'B-transport_type', 'B-radio_name', 'I-change_amount', 'B-transport_agency', 'B-movie_name', 'B-general_frequency', 'B-event_name', 'I-joke_type', 'B-relation', 'B-coffee_type', 'I-order_type', 'B-email_address', 'B-app_name', 'B-podcast_descriptor', 'B-definition_word', 'I-list_name', 'B-audiobook_author', 'I-audiobook_author', 'I-sport_type', 'I-news_topic', 'B-personal_info', 'I-radio_name', 'B-person', 'I-color_type', 'I-food_type', 'I-device_type', 'B-time_zone', 'I-music_genre', 'B-business_type', 'B-currency_name', 'I-currency_name', 'B-ingredient', 'I-transport_name', 'B-game_name', 'I-time_zone']
+
 class MASSIVE(datasets.GeneratorBasedBuilder):
     """MASSIVE: A 1M-Example Multilingual Natural Language Understanding Dataset with 51 Typologically-Diverse Languages"""
 
@@ -285,10 +287,12 @@ class MASSIVE(datasets.GeneratorBasedBuilder):
                     "intent": datasets.features.ClassLabel(names=_INTENTS),
                     "utt": datasets.Value("string"),
                     "annot_utt": datasets.Value("string"),
-                    "annot_utt_bio": datasets.Sequence({
-                        "word": datasets.Value("string"),
-                        "tag": datasets.Value("string"),
-                    }),
+                    "tokens": datasets.Sequence(datasets.Value("string")),
+                    "ner_tags": datasets.Sequence(
+                        datasets.features.ClassLabel(
+                            names = _TAGS
+                        )
+                    ),
                     "worker_id": datasets.Value("string"),
                     "slot_method": datasets.Sequence({
                         "slot": datasets.Value("string"),
@@ -343,7 +347,7 @@ class MASSIVE(datasets.GeneratorBasedBuilder):
 
     def _getBioFormat(self, text):
 
-        tags, words = [], []
+        tags, tokens = [], []
 
         bio_mode = False
         cpt_bio = 0
@@ -386,9 +390,9 @@ class MASSIVE(datasets.GeneratorBasedBuilder):
                     word = s
             
             tags.append(token)
-            words.append(word)
+            tokens.append(word)
 
-        return [{"word": w, "tag": t} for w, t in zip(words, tags)]
+        return tokens, tags
 
     def _generate_examples(self, filepath, split, lang):
 
@@ -436,6 +440,8 @@ class MASSIVE(datasets.GeneratorBasedBuilder):
             else:
                 judgments = []
 
+            tokens, tags = self._getBioFormat(data["annot_utt"])
+
             yield key_, {
                 "id": data["id"],
                 "locale": data["locale"],
@@ -444,7 +450,8 @@ class MASSIVE(datasets.GeneratorBasedBuilder):
                 "intent": data["intent"],
                 "utt": data["utt"],
                 "annot_utt": data["annot_utt"],
-                "annot_utt_bio": self._getBioFormat(data["annot_utt"]),
+                "tokens": tokens,
+                "ner_tags": tags,
                 "worker_id": data["worker_id"],
                 "slot_method": slot_method,
                 "judgments": judgments,

README.md

@@ -265,19 +265,17 @@ print(dataset[0])
   "intent": 48,
   "utt": "réveille-moi à neuf heures du matin le vendredi",
   "annot_utt": "réveille-moi à [time : neuf heures du matin] le [date : vendredi]",
-  "annot_utt_bio": {
-    "word": [
-      "réveille-moi",
-      "à",
-      "neuf",
-      "heures",
-      "du",
-      "matin",
-      "le",
-      "vendredi"
-    ],
-    "tag": ["O", "O", "B-time", "I-time", "I-time", "I-time", "O", "B-date"]
-  },
+  "tokens": [
+    "réveille-moi",
+    "à",
+    "neuf",
+    "heures",
+    "du",
+    "matin",
+    "le",
+    "vendredi"
+  ],
+  "ner_tags": [44, 44, 71, 5, 5, 5, 44, 13],
   "worker_id": "22",
   "slot_method": {
     "slot": ["time", "date"],