Merge pull request #162 from vthorsteinsson/ice

Bug fixes in inference and data generation; faster token unescaping

tensor2tensor/data_generators/generator_utils.py

@@ -329,18 +329,19 @@ def get_or_generate_tabbed_vocab(tmp_dir, source_filename,
     return vocab
 
   # Use Tokenizer to count the word occurrences.
+  token_counts = defaultdict(int)
   filepath = os.path.join(tmp_dir, source_filename)
   with tf.gfile.GFile(filepath, mode="r") as source_file:
     for line in source_file:
       line = line.strip()
       if line and "\t" in line:
         parts = line.split("\t", maxsplit=1)
         part = parts[index].strip()
-        _ = tokenizer.encode(text_encoder.native_to_unicode(part))
+        for tok in tokenizer.encode(text_encoder.native_to_unicode(part)):
+          token_counts[tok] += 1
 
   vocab = text_encoder.SubwordTextEncoder.build_to_target_size(
-      vocab_size, tokenizer.token_counts, 1,
-      min(1e3, vocab_size + text_encoder.NUM_RESERVED_TOKENS))
+      vocab_size, token_counts, 1, 1e3)
   vocab.store_to_file(vocab_filepath)
   return vocab
 

tensor2tensor/data_generators/text_encoder.py

@@ -24,6 +24,7 @@
 from __future__ import print_function
 
 from collections import defaultdict
+import re
 
 # Dependency imports
 
@@ -225,6 +226,7 @@ class SubwordTextEncoder(TextEncoder):
 
   def __init__(self, filename=None):
     """Initialize and read from a file, if provided."""
+    self._alphabet = set()
     if filename is not None:
       self._load_from_file(filename)
     super(SubwordTextEncoder, self).__init__(num_reserved_ids=None)
@@ -503,6 +505,12 @@ def _escape_token(self, token):
         ret += u"\\%d;" % ord(c)
     return ret
 
+  # Regular expression for unescaping token strings
+  # '\u' is converted to '_'
+  # '\\' is converted to '\'
+  # '\213;' is converted to unichr(213)
+  _UNESCAPE_REGEX = re.compile(u'|'.join([r"\\u", r"\\\\", r"\\([0-9]+);"]))
+
   def _unescape_token(self, escaped_token):
     """Inverse of _escape_token().
 
@@ -511,32 +519,14 @@ def _unescape_token(self, escaped_token):
     Returns:
       token: a unicode string
     """
-    ret = u""
-    escaped_token = escaped_token[:-1]
-    pos = 0
-    while pos < len(escaped_token):
-      c = escaped_token[pos]
-      if c == "\\":
-        pos += 1
-        if pos >= len(escaped_token):
-          break
-        c = escaped_token[pos]
-        if c == u"u":
-          ret += u"_"
-          pos += 1
-        elif c == "\\":
-          ret += u"\\"
-          pos += 1
-        else:
-          semicolon_pos = escaped_token.find(u";", pos)
-          if semicolon_pos == -1:
-            continue
-          try:
-            ret += unichr(int(escaped_token[pos:semicolon_pos]))
-            pos = semicolon_pos + 1
-          except (ValueError, OverflowError) as _:
-            pass
-      else:
-        ret += c
-        pos += 1
-    return ret
+    def match(m):
+      if m.group(1) is not None:
+        # Convert '\213;' to unichr(213)
+        try:
+          return unichr(int(m.group(1)))
+        except (ValueError, OverflowError) as _:
+          return ""
+      # Convert '\u' to '_' and '\\' to '\'
+      return u"_" if m.group(0) == u"\\u" else u"\\"
+    # Cut off the trailing underscore and apply the regex substitution
+    return self._UNESCAPE_REGEX.sub(match, escaped_token[:-1])

tensor2tensor/data_generators/tokenizer_test.py

@@ -1,3 +1,4 @@
+# -*- coding: utf-8 -*-
 # Copyright 2017 The Tensor2Tensor Authors.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
@@ -12,7 +13,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-# coding=utf-8
 """Tests for tensor2tensor.data_generators.tokenizer."""
 
 from __future__ import absolute_import

tensor2tensor/utils/trainer_utils.py

@@ -585,6 +585,7 @@ def decode_from_dataset(estimator):
     tf.logging.info("Performing local inference.")
     infer_problems_data = get_datasets_for_mode(hparams.data_dir,
                                                 tf.contrib.learn.ModeKeys.INFER)
+
     infer_input_fn = get_input_fn(
         mode=tf.contrib.learn.ModeKeys.INFER,
         hparams=hparams,
@@ -625,9 +626,11 @@ def log_fn(inputs,
 
     # The function predict() returns an iterable over the network's
     # predictions from the test input. We use it to log inputs and decodes.
-    for j, result in enumerate(result_iter):
-      inputs, targets, outputs = (result["inputs"], result["targets"],
-                                  result["outputs"])
+    inputs_iter = result_iter["inputs"]
+    targets_iter = result_iter["targets"]
+    outputs_iter = result_iter["outputs"]
+    for j, result in enumerate(zip(inputs_iter, targets_iter, outputs_iter)):
+      inputs, targets, outputs = result
       if FLAGS.decode_return_beams:
         output_beams = np.split(outputs, FLAGS.decode_beam_size, axis=0)
         for k, beam in enumerate(output_beams):