Cast regularization parameters to float. (tensorflow#1707)

This works around a bug in earlier proto versions
that automatically infer these values to be integer
instead of float.

object_detection/builders/hyperparams_builder.py

@@ -111,9 +111,9 @@ def _build_regularizer(regularizer):
  """
  regularizer_oneof = regularizer.WhichOneof('regularizer_oneof')
  if regularizer_oneof == 'l1_regularizer':
- return slim.l1_regularizer(scale=regularizer.l1_regularizer.weight)
+ return slim.l1_regularizer(scale=float(regularizer.l1_regularizer.weight))
  if regularizer_oneof == 'l2_regularizer':
- return slim.l2_regularizer(scale=regularizer.l2_regularizer.weight)
+ return slim.l2_regularizer(scale=float(regularizer.l2_regularizer.weight))
  raise ValueError('Unknown regularizer function: {}'.format(regularizer_oneof))