Add usage of cpp predict client and remove generate files

Author: tobegit3hub

README.md

@@ -16,9 +16,10 @@ Following are the supported features.
   - [x] [Python gPRC client](./python_predict_client/)
   - [x] [Java gPRC client](./java_predict_client/)
   - [x] [Scala gPRC client](./java_predict_client/)
-  - [x] [Golang client](./golang_predict_client/)
+  - [x] [Golang gRPC client](./golang_predict_client/)
+  - [x] [C++ gRPC client](./cpp_predict_client/)
   - [ ] Spark client
-  - [ ] C++ client
+  - [ ] Android client
 - [x] Use Cases
   - [x] Train model
   - [x] Export model
@@ -38,6 +39,7 @@ Following are the supported features.
   - [x] Optimizers
   - [x] Learning rate decay
   - [x] Batch normalization
+  - [ ] Distributed training
 
 ## Usage
 

cpp_predict_client/README.md

@@ -1,16 +1,35 @@
-# Python Predict Client
+# Cpp Predict Client
 
 ## Introduction
 
 TensorFlow serving is the gRPC service for general TensorFlow models. We can implement the C++ gRPC client to predict.
 
+If you are using `bazel`, refer to to [inception_client.cc](https://github.com/tensorflow/serving/pull/300)
+
 ## Usage
 
-1. Export the TensorFlow models.
-2. Run with `./predict_client.py --host 127.0.0.1 --port 9000 --model_name cancer --model_version 1`
+Add the binary in `tensorflow_serving/example/BUILD`.
+
+```
+cc_binary(
+    name = "tensorflow_model_client",
+    srcs = [
+        "serving_client_cpp.cc",
+    ],
+    deps = [
+        "//tensorflow_serving/apis:prediction_service_proto",
+    ],
+)
+```
+
+Compile the project.
 
-## Development
+```
+bazel build //tensorflow_serving/example:tensorflow_model_client
+```
 
-Build protobuf from source and refer to <https://github.com/google/protobuf/blob/master/src/README.md>.
+Run the predict client.
 
-Generetate code from proto files.
+```
+bazel-bin/tensorflow_serving/example/tensorflow_model_client
+```