Address PR comments.

Author: asimshankar

README.md

@@ -3,13 +3,13 @@
 These examples include using pre-trained models for [image
 classification](label_image) and [object detection](object_detection),
 and driving the [training](training) of a pre-defined model - all using the
-TensorFlow Java APIs.
+TensorFlow Java API.
 
 The TensorFlow Java API does not have feature parity with the Python API.
-The Java APIs are most suitable for inference using pre-trained models
+The Java API is most suitable for inference using pre-trained models
 and for training pre-defined models from a single Java process.
 
-However, Python will be the most convenient language for defining the
+Python will be the most convenient language for defining the
 numerical computation of a model.
 
 - [Slides](https://docs.google.com/presentation/d/e/2PACX-1vQ6DzxNTBrJo7K5P8t5_rBRGnyJoPUPBVOJR4ooHCwi4TlBFnIriFmI719rDNpcQzojqsV58aUqmBBx/pub?start=false&loop=false&delayms=3000) from January 2018.