Add TPU SavedModel exporter and refactor OD code (#6737)

247226201  by ronnyvotel:

    Updating the visualization tools to accept unique_ids for color coding.

--
247067830  by Zhichao Lu:

    Add box_encodings_clip_range options for the convolutional box predictor (for TPU compatibility).

--
246888475  by Zhichao Lu:

    Remove unused _update_eval_steps function.

--
246163259  by lzc:

    Add a gather op that can handle ignore indices (which are "-1"s in this case).

--
246084944  by Zhichao Lu:

    Keras based implementation for SSD + MobilenetV2 + FPN.

--
245544227  by rathodv:

    Add batch_get_targets method to target assigner module to gather any groundtruth tensors based on the results of target assigner.

--
245540854  by rathodv:

    Update target assigner to return match tensor instead of a match object.

--
245434441  by Zhichao Lu:

    Add README for tpu_exporters package.

--
245381834  by lzc:

    Internal change.

--
245298983  by Zh...

research/object_detection/predictors/convolutional_box_predictor.py

@@ -357,9 +357,9 @@ class WeightSharedConvolutionalBoxPredictor(box_predictor.BoxPredictor):
       inserted_layer_counter = 0
       target_channel = max(set(feature_channels), key=feature_channels.count)
       tf.logging.info('Not all feature maps have the same number of '
-                      'channels, found: {}, addition project layers '
-                      'to bring all feature maps to uniform channels '
-                      'of {}'.format(feature_channels, target_channel))
+                      'channels, found: {}, appending additional projection '
+                      'layers to bring all feature maps to uniformly have {} '
+                      'channels.'.format(feature_channels, target_channel))
     else:
       # Place holder variables if has_different_feature_channels is False.
       target_channel = -1
@@ -377,6 +377,8 @@ class WeightSharedConvolutionalBoxPredictor(box_predictor.BoxPredictor):
       with tf.variable_scope('WeightSharedConvolutionalBoxPredictor',
                              reuse=tf.AUTO_REUSE):
         with slim.arg_scope(self._conv_hyperparams_fn()):
+          # TODO(wangjiang) Pass is_training to the head class directly.
+          with slim.arg_scope([slim.dropout], is_training=self._is_training):
             (image_feature,
              inserted_layer_counter) = self._insert_additional_projection_layer(
                  image_feature, inserted_layer_counter, target_channel)