Add docker file and the necessary documentation

docker/Dockerfile

@@ -0,0 +1,54 @@
+FROM ubuntu:16.04
+USER root
+LABEL maintainer="github.com/fai555"
+
+RUN apt-get update && \
+        apt-get install -y software-properties-common && \
+        add-apt-repository ppa:deadsnakes/ppa && \
+        apt-get update
+
+RUN apt-get install -y build-essential python3.6 python3.6-dev python3-pip python3.6-venv
+RUN python3.6 -m pip install pip --upgrade
+
+# Pick up some TF dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+        build-essential \
+        curl \
+        libfreetype6-dev \
+        libpng12-dev \
+        libzmq3-dev \
+        pkg-config \
+        rsync \
+        git \
+        software-properties-common \
+        unzip \
+        wget \
+        && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+
+RUN pip3 --no-cache-dir install \
+        tensorflow==1.13.2
+
+RUN pip3 --no-cache-dir install \
+        Cython \
+        contextlib2 \
+        jupyter \
+        matplotlib \
+        pillow \
+        lxml \
+        wheel \
+        pandas
+
+RUN pip3 install "git+https://github.com/philferriere/cocoapi.git#egg=pycocotools&subdirectory=PythonAPI"
+WORKDIR /app
+RUN git clone https://github.com/fai555/tensorflow-serving_sidecar.git
+WORKDIR /app/tensorflow-serving_sidecar
+RUN pip3 install -r requirements.txt
+RUN wget -O protobuf.zip https://github.com/google/protobuf/releases/download/v3.9.0/protoc-3.9.0-linux-x86_64.zip && \
+    unzip protobuf.zip && \
+    ./bin/protoc object_detection/protos/*.proto --python_out=.
+
+ENV PYTHONPATH "/app/models/research:/app/models/research/slim:${PYTHONPATH}"
+ENTRYPOINT ["python3.6"]

docs/docker_prediction_request_client.md

@@ -0,0 +1,38 @@
+# Build Docker Images
+
+You can build your own docker image using the Dockerfile in the docker direcoty.
+```
+docker build -t tensorflow-serving-sidecar-client:latest .
+```
+
+# Public docker image
+
+A public docker image in Google Cloud Container Registry is availabe if you don't want to build the image yourself. You can just pull the image and use.
+```
+asia.gcr.io/im-mlpipeline/tensorflow-serving-sidecar-client:latest
+```
+
+
+
+
+
+# Sending Prediction Requests
+You can use the following command to send prediction requests.
+
+```
+# absolute path to the directory where label_map.pbtxt and input image files are. It must be an absolute path. Docker volume mounting doesn't work with relative path. All other paths in this scripts will work with relative path.
+export VOLUME_PATH=""
+export SERVER_URL="http://34.73.137.228:8501/v1/models/faster_rcnn_resnet:predict"
+# relative path to the test image. i.e. config/image1.jpg. The path is relative to VOLUME_PATH
+export IMAGE_PATH=""
+# relative path to the output json. i.e. config/out_image1.json. The path is relative to VOLUME_PATH
+export OUTPUT_JSON=""
+# relative path to the label_map.pbtxt. i.e. config/label_map.pbtxt. The path is relative to VOLUME_PATH
+export LABEL_MAP=""
+# Specify True if you want to save the output image. 
+export SAVE_OUTPUT_IMAGE=True
+# You can use the publicly available docker image or use your own image
+export DOCKER_IMAGE_NAME="asia.gcr.io/im-mlpipeline/tensorflow-serving-sidecar-client:latest"
+
+docker run -it -v ${VOLUME_PATH}:/app/tensorflow-serving_sidecar/config -t ${DOCKER_IMAGE_NAME} client.py --server_url=${SERVER_URL}  --image_path=${IMAGE_PATH} --output_json=${OUTPUT_JSON} --save_output_image=${SAVE_OUTPUT_IMAGE} --label_map=${LABEL_MAP}
+```