Docker image in doc site (#1117)

* update * Update index.md * Update index.md * Update README.md
intel-analytics · Jan 24, 2019 · 59006b7 · 59006b7
1 parent 212a894
commit 59006b7
Show file tree

Hide file tree

Showing 2 changed files with 299 additions and 6 deletions.
diff --git a/README.md b/README.md
@@ -56,6 +56,12 @@ In addition, Analytics Zoo also provides a rich set of analytics and AI support
 
 - [Reference use cases](#reference-use-cases): a collection of end-to-end *reference use cases* (e.g., anomaly detection, sentiment analysis, fraud detection, image augmentation, object detection, variational autoencoder, etc.)
 
+- [Docker images and builders](#docker-images-and-builders)
+    - [Analytics-Zoo in Docker](#analytics-zoo-in-docker)
+    - [How to build it](#how-to-build-it)
+    - [How to use the image](#how-to-use-the-image)
+    - [Notice](#notice)
+
 ## _Distributed TensorFlow and Keras on Spark/BigDL_
 To make it easy to build and productionize the deep learning applications for Big Data, Analytics Zoo provides a unified analytics + AI platform that seamlessly unites Spark, TensorFlow, Keras and BigDL programs into an integrated pipeline (as illustrated below), which can then transparently run on a large-scale Hadoop/Spark clusters for distributed training and inference. (Please see more details [here](https://analytics-zoo.github.io/master/#ProgrammingGuide/tensorflow/)).
 
@@ -301,3 +307,148 @@ Using *Analytics Zoo Image Classification API* (including a set of pretrained de
 
 ## _Reference use cases_
 Analytics Zoo provides a collection of end-to-end reference use cases, including *time series anomaly detection*, *sentiment analysis*, *fraud detection*, *image similarity*, etc. (See more details [here](https://analytics-zoo.github.io/master/#ProgrammingGuide/usercases-overview/))
+
+## _Docker images and builders_
+
+### _Analytics-Zoo in Docker_
+
+**By default, the Analytics-Zoo image has installed below packages:**
+- git
+- maven
+- Oracle jdk 1.8.0_152 (in /opt/jdk1.8.0_152)
+- python 2.7.6
+- pip
+- numpy
+- scipy
+- pandas
+- scikit-learn
+- matplotlib
+- seaborn
+- jupyter
+- wordcloud
+- moviepy
+- requests
+- tensorflow_
+- spark-${SPARK_VERSION} (in /opt/work/spark-${SPARK_VERSION})
+- Analytics-Zoo distribution (in /opt/work/analytics-zoo-${ANALYTICS_ZOO_VERSION})
+- Analytics-Zoo source code (in /opt/work/analytics-zoo)
+
+**The work dir for Analytics-Zoo is /opt/work.**
+- download-analytics-zoo.sh is used for downloading Analytics-Zoo distributions.
+- start-notebook.sh is used for starting the jupyter notebook. You can specify the environment settings and spark settings to start a specified jupyter notebook.
+- analytics-Zoo-${ANALYTICS_ZOO_VERSION} is the Analytics-Zoo home of Analytics-Zoo distribution.
+- analytics-zoo-SPARK_x.x-x.x.x-dist.zip is the zip file of Analytics-Zoo distribution.
+- spark-${SPARK_VERSION} is the Spark home.
+- analytics-zoo is cloned from https://github.com/intel-analytics/analytics-zoo, contains apps, examples using analytics-zoo.
+
+### _How to build it_
+
+**By default, you can build a Analytics-Zoo:default image with latest nightly-build Analytics-Zoo distributions:**
+
+```bash
+sudo docker build --rm -t intelanalytics/analytics-zoo:default .
+```
+
+**If you need http and https proxy to build the image:**
+```bash
+sudo docker build \
+    --build-arg http_proxy=http://your-proxy-host:your-proxy-port \
+    --build-arg https_proxy=https://your-proxy-host:your-proxy-port \
+    --rm -t intelanalytics/analytics-zoo:default .
+```
+
+**You can also specify the ANALYTICS_ZOO_VERSION and SPARK_VERSION to build a specific Analytics-Zoo image:**
+```bash
+sudo docker build \
+    --build-arg http_proxy=http://your-proxy-host:your-proxy-port \
+    --build-arg https_proxy=https://your-proxy-host:your-proxy-port \
+    --build-arg ANALYTICS_ZOO_VERSION=0.3.0 \
+    --build-arg BIGDL_VERSION=0.6.0 \
+    --build-arg SPARK_VERSION=2.3.1 \
+    --rm -t intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1 .
+```
+
+### _How to use the image_
+**To start a notebook directly with a specified port(e.g. 12345). You can view the notebook on http://[host-ip]:12345**
+```bash
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+```
+
+**If you need http and https proxy in your environment:**
+```bash
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:default
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:default
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+```
+
+**You can also start the container first**
+```bash
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default bash
+```
+
+**In the container, after setting proxy and ports, you can start the Notebook by:**
+```bash
+/opt/work/start-notebook.sh
+```
+
+### _Notice_
+**If you need nightly build version of Analytics-Zoo, please pull the image form Dockerhub with:**
+```bash
+sudo docker pull intelanalytics/analytics-zoo:latest
+```
+
+**Please follow the readme in each app folder to test the jupyter notebooks !!!**
+
+**With 0.3+ version of Anaytics-Zoo Docker image, you can specify the runtime conf of spark**
+```bash
+sudo docker run -itd --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="1234qwer" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port  \
+    -e https_proxy=https://your-proxy-host:your-proxy-port  \
+    -e RUNTIME_DRIVER_CORES_ENV=4 \
+    -e RUNTIME_DRIVER_MEMORY=20g \
+    -e RUNTIME_EXECUTOR_CORES=4 \
+    -e RUNTIME_EXECUTOR_MEMORY=20g \
+    -e RUNTIME_TOTAL_EXECUTOR_CORES=4 \
+    intelanalytics/analytics-zoo:latest
+```
diff --git a/docs/docs/index.md b/docs/docs/index.md
@@ -56,6 +56,12 @@ In addition, Analytics Zoo also provides a rich set of analytics and AI support
 
 - [Reference use cases](#reference-use-cases): a collection of end-to-end *reference use cases* (e.g., anomaly detection, sentiment analysis, fraud detection, image augmentation, object detection, variational autoencoder, etc.)
 
+- [Docker images and builders](#docker-images-and-builders)
+    - [Analytics-Zoo in Docker](#analytics-zoo-in-docker)
+    - [How to build it](#how-to-build-it)
+    - [How to use the image](#how-to-use-the-image)
+    - [Notice](#notice)
+
 ## _Distributed TensorFlow and Keras on Spark/BigDL_
 To make it easy to build and productionize the deep learning applications for Big Data, Analytics Zoo provides a unified analytics + AI platform that seamlessly unites Spark, TensorFlow, Keras and BigDL programs into an integrated pipeline (as illustrated below), which can then transparently run on a large-scale Hadoop/Spark clusters for distributed training and inference. (Please see more details [here](https://analytics-zoo.github.io/master/#ProgrammingGuide/tensorflow/)).
 
@@ -303,20 +309,156 @@ Using *Analytics Zoo Image Classification API* (including a set of pretrained de
 ## _Reference use cases_
 Analytics Zoo provides a collection of end-to-end reference use cases, including *time series anomaly detection*, *sentiment analysis*, *fraud detection*, *image similarity*, etc. (See more details [here](https://analytics-zoo.github.io/master/#ProgrammingGuide/usercases-overview/))
 
+## _Docker images and builders_
+
+### _Analytics-Zoo in Docker_
+
+**By default, the Analytics-Zoo image has installed below packages:**
+- git
+- maven
+- Oracle jdk 1.8.0_152 (in /opt/jdk1.8.0_152)
+- python 2.7.6
+- pip
+- numpy
+- scipy
+- pandas
+- scikit-learn
+- matplotlib
+- seaborn
+- jupyter
+- wordcloud
+- moviepy
+- requests
+- tensorflow_
+- spark-${SPARK_VERSION} (in /opt/work/spark-${SPARK_VERSION})
+- Analytics-Zoo distribution (in /opt/work/analytics-zoo-${ANALYTICS_ZOO_VERSION})
+- Analytics-Zoo source code (in /opt/work/analytics-zoo)
+
+**The work dir for Analytics-Zoo is /opt/work.**
+- download-analytics-zoo.sh is used for downloading Analytics-Zoo distributions.
+- start-notebook.sh is used for starting the jupyter notebook. You can specify the environment settings and spark settings to start a specified jupyter notebook.
+- analytics-Zoo-${ANALYTICS_ZOO_VERSION} is the Analytics-Zoo home of Analytics-Zoo distribution.
+- analytics-zoo-SPARK_x.x-x.x.x-dist.zip is the zip file of Analytics-Zoo distribution.
+- spark-${SPARK_VERSION} is the Spark home.
+- analytics-zoo is cloned from https://github.com/intel-analytics/analytics-zoo, contains apps, examples using analytics-zoo.
+
+### _How to build it_
+
+**By default, you can build a Analytics-Zoo:default image with latest nightly-build Analytics-Zoo distributions:**
+
+```bash
+sudo docker build --rm -t intelanalytics/analytics-zoo:default .
+```
 
+**If you need http and https proxy to build the image:**
+```bash
+sudo docker build \
+    --build-arg http_proxy=http://your-proxy-host:your-proxy-port \
+    --build-arg https_proxy=https://your-proxy-host:your-proxy-port \
+    --rm -t intelanalytics/analytics-zoo:default .
+```
 
+**You can also specify the ANALYTICS_ZOO_VERSION and SPARK_VERSION to build a specific Analytics-Zoo image:**
+```bash
+sudo docker build \
+    --build-arg http_proxy=http://your-proxy-host:your-proxy-port \
+    --build-arg https_proxy=https://your-proxy-host:your-proxy-port \
+    --build-arg ANALYTICS_ZOO_VERSION=0.3.0 \
+    --build-arg BIGDL_VERSION=0.6.0 \
+    --build-arg SPARK_VERSION=2.3.1 \
+    --rm -t intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1 .
+```
 
+### _How to use the image_
+**To start a notebook directly with a specified port(e.g. 12345). You can view the notebook on http://[host-ip]:12345**
+```bash
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default
+
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default
+
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+```
 
+**If you need http and https proxy in your environment:**
+```bash
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:default
+
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:default
+
+sudo docker run -it --rm -p 12345:12345 \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port \
+    -e https_proxy=https://your-proxy-host:your-proxy-port \
+    intelanalytics/analytics-zoo:0.3.0-bigdl_0.6.0-spark_2.3.1
+```
 
+**You can also start the container first**
+```bash
+sudo docker run -it --rm --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="your-token" \
+    intelanalytics/analytics-zoo:default bash
+```
 
+**In the container, after setting proxy and ports, you can start the Notebook by:**
+```bash
+/opt/work/start-notebook.sh
+```
 
+### _Notice_
+**If you need nightly build version of Analytics-Zoo, please pull the image form Dockerhub with:**
+```bash
+sudo docker pull intelanalytics/analytics-zoo:latest
+```
 
-
-
-
-
-
-
+**Please follow the readme in each app folder to test the jupyter notebooks !!!**
+
+**With 0.3+ version of Anaytics-Zoo Docker image, you can specify the runtime conf of spark**
+```bash
+sudo docker run -itd --net=host \
+    -e NotebookPort=12345 \
+    -e NotebookToken="1234qwer" \
+    -e http_proxy=http://your-proxy-host:your-proxy-port  \
+    -e https_proxy=https://your-proxy-host:your-proxy-port  \
+    -e RUNTIME_DRIVER_CORES_ENV=4 \
+    -e RUNTIME_DRIVER_MEMORY=20g \
+    -e RUNTIME_EXECUTOR_CORES=4 \
+    -e RUNTIME_EXECUTOR_MEMORY=20g \
+    -e RUNTIME_TOTAL_EXECUTOR_CORES=4 \
+    intelanalytics/analytics-zoo:latest
+```