Merge remote-tracking branch 'upstream/master'

examples/extract_features.py

@@ -57,7 +57,7 @@ class InputFeatures(object):
 
 
 def convert_examples_to_features(examples, seq_length, tokenizer):
-    """Loads a data file into a list of `InputBatch`s."""
+    """Loads a data file into a list of `InputFeature`s."""
 
     features = []
     for (ex_index, example) in enumerate(examples):

examples/lm_finetuning/pregenerate_training_data.py

@@ -49,7 +49,7 @@ class DocumentDatabase:
                 self._precalculate_doc_weights()
             rand_start = self.doc_cumsum[current_idx]
             rand_end = rand_start + self.cumsum_max - self.doc_lengths[current_idx]
-            sentence_index = randint(rand_start, rand_end) % self.cumsum_max
+            sentence_index = randint(rand_start, rand_end-1) % self.cumsum_max
             sampled_doc_index = np.searchsorted(self.doc_cumsum, sentence_index, side='right')
         else:
             # If we don't use sentence weighting, then every doc has an equal chance to be chosen

examples/run_classifier.py

@@ -442,7 +442,7 @@ def convert_examples_to_features(examples, label_list, max_seq_length,
         # sequence or the second sequence. The embedding vectors for `type=0` and
         # `type=1` were learned during pre-training and are added to the wordpiece
         # embedding vector (and position vector). This is not *strictly* necessary
-        # since the [SEP] token unambigiously separates the sequences, but it makes
+        # since the [SEP] token unambiguously separates the sequences, but it makes
         # it easier for the model to learn the concept of sequences.
         #
         # For classification tasks, the first vector (corresponding to [CLS]) is

examples/run_squad.py

@@ -85,9 +85,9 @@ class SquadExample(object):
         s += ", doc_tokens: [%s]" % (" ".join(self.doc_tokens))
         if self.start_position:
             s += ", start_position: %d" % (self.start_position)
-        if self.start_position:
+        if self.end_position:
             s += ", end_position: %d" % (self.end_position)
-        if self.start_position:
+        if self.is_impossible:
             s += ", is_impossible: %r" % (self.is_impossible)
         return s