Fix for string features (tensorflow#1440)

Author: epurdy

tensor2tensor/utils/t2t_model.py

@@ -2017,7 +2017,7 @@ def summarize_features(features, num_shards=1):
 
   with tf.name_scope("input_stats"):
     for (k, v) in sorted(six.iteritems(features)):
-      if isinstance(v, tf.Tensor) and v.get_shape().ndims > 1:
+      if isinstance(v, tf.Tensor) and v.get_shape().ndims > 1 and v.dtype != tf.string:
         tf.summary.scalar("%s_batch" % k, tf.shape(v)[0] // num_shards)
         tf.summary.scalar("%s_length" % k, tf.shape(v)[1])
         nonpadding = tf.to_float(tf.not_equal(v, 0))