Update indian_names.py

indian_names.py

@@ -56,7 +56,8 @@ class indina_names(datasets.GeneratorBasedBuilder):
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}),
         ]
 
-     def _generate_examples(self, filepath):
+       def _generate_examples(self, filepath):
+        logger.info("⏳ Generating examples from = %s", filepath)
         with open(filepath, encoding="utf-8") as f:
             current_tokens = []
             current_labels = []
@@ -64,13 +65,15 @@ class indina_names(datasets.GeneratorBasedBuilder):
             for row in f:
                 row = row.rstrip()
                 if row:
-                    token, label = row.split(",")
+                    token, label = row.split("\t")
                     current_tokens.append(token)
                     current_labels.append(label)
                 else:
+                    # New sentence
                     if not current_tokens:
+                        # Consecutive empty lines will cause empty sentences
                         continue
-                    assert len(current_tokens) == len(current_labels), "Mismatch between tokens and labels"
+                    assert len(current_tokens) == len(current_labels), "💔 between len of tokens & labels"
                     sentence = (
                         sentence_counter,
                         {
@@ -83,6 +86,7 @@ class indina_names(datasets.GeneratorBasedBuilder):
                     current_tokens = []
                     current_labels = []
                     yield sentence
+            # Don't forget last sentence in dataset 🧐
             if current_tokens:
                 yield sentence_counter, {
                     "id": str(sentence_counter),