debug

bernice-pretrain-data.py

@@ -67,6 +67,7 @@ _URLS = {
     "indic": ["indic_tweet_ids.txt.gz"]
 }
 
+
 # TODO: Name of the dataset usually matches the script name with CamelCase instead of snake_case
 class BernicePretrainData(datasets.GeneratorBasedBuilder):
     """Tweet IDs for the 2.5 billion multilingual tweets used to train Bernice, a Twitter encoder."""
@@ -141,7 +142,7 @@ class BernicePretrainData(datasets.GeneratorBasedBuilder):
         # TODO: This method handles input defined in _split_generators to yield (key, example) tuples from the dataset.
         # The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
         for filepath in filepaths:
-            with gzip.open(filepath, "rb") as f:
+            with open(filepath, encoding="utf-8") as f:
                 for line_number, instance in enumerate(f):
                     tweet_id, lang, year = instance.strip().split("\t")
                     yield tweet_id, {