From 0fb11167547eb109064f2c687554a3709bc65312 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin.zuo@sjtu.edu.cn>
Date: Wed, 17 Jan 2018 16:38:24 +0800
Subject: [PATCH 01/16] Add toree installation script for Scala&PySpark kernel.

---
 toree_install.sh | 51 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 51 insertions(+)
 create mode 100755 toree_install.sh

diff --git a/toree_install.sh b/toree_install.sh
new file mode 100755
index 0000000..f5eb80c
--- /dev/null
+++ b/toree_install.sh
@@ -0,0 +1,51 @@
+#!/bin/bash
+
+# Find the path of BigDL and Spark
+export BIGDL_PIP_HOME=`pip show BigDL | sed -n -e '/^Location/p' | sed 's/[^ ]* //'`
+export BIGDL_HOME=${BIGDL_PIP_HOME}/bigdl/share
+export PYSPARK_PIP_HOME=`pip show pyspark | sed -n -e '/^Location/p' | sed 's/[^ ]* //'`
+export SPARK_HOME=`python ${PYSPARK_PIP_HOME}/pyspark/find_spark_home.py`
+# export SPARK_HOME=${SPARK_PATH}/bin
+
+# Check installation of BigDL
+if [ -z "${BIGDL_HOME}" ]; then
+	echo "Please install BigDL correctly!"
+	exit 1
+fi
+
+# Check installation of Spark
+if [ -z "${SPARK_HOME}" ]; then
+	echo "Please set install Spark correctly!"
+	exit 1
+fi
+
+# Set paths
+export BIGDL_JAR_NAME=`ls ${BIGDL_HOME}/lib/ | grep jar-with-dependencies.jar`
+export BIGDL_JAR=${BIGDL_HOME}/lib/${BIGDL_JAR_NAME}
+export BIGDL_PY_ZIP_NAME=`ls ${BIGDL_HOME}/lib/ | grep python-api.zip`
+export BIGDL_PY_ZIP=${BIGDL_HOME}/lib/${BIGDL_PY_ZIP_NAME}
+export BIGDL_CONF=${BIGDL_HOME}/conf/spark-bigdl.conf
+echo ${BIGDL_JAR}
+echo ${BIGDL_PY_ZIP}
+echo ${BIGDL_CONF}
+# Check files
+if [ ! -f ${BIGDL_JAR} ]; then
+    echo "Cannot find ${BIGDL_JAR}!"
+    exit 1
+fi
+
+if [ ! -f ${BIGDL_PY_ZIP} ]; then
+    echo "Cannot find ${BIGDL_PY_ZIP}!"
+    exit 1
+fi
+
+if [ ! -f ${BIGDL_CONF} ]; then
+    echo "Cannot find ${BIGDL_CONF}!"
+    exit 1
+fi
+
+# Configure Spark
+export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory'"
+
+# Install Toree
+jupyter toree install --interpreters=Scala,PySpark --spark_home=${SPARK_HOME} --spark_opts='${SPARK_OPTS}'

From e3b482e36195be9d179ebf8d77d3f51981e41bf0 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin.zuo@sjtu.edu.cn>
Date: Wed, 17 Jan 2018 16:43:39 +0800
Subject: [PATCH 02/16] Refine the script.

---
 toree_install.sh | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/toree_install.sh b/toree_install.sh
index f5eb80c..1db0fc5 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -5,7 +5,6 @@ export BIGDL_PIP_HOME=`pip show BigDL | sed -n -e '/^Location/p' | sed 's/[^ ]*
 export BIGDL_HOME=${BIGDL_PIP_HOME}/bigdl/share
 export PYSPARK_PIP_HOME=`pip show pyspark | sed -n -e '/^Location/p' | sed 's/[^ ]* //'`
 export SPARK_HOME=`python ${PYSPARK_PIP_HOME}/pyspark/find_spark_home.py`
-# export SPARK_HOME=${SPARK_PATH}/bin
 
 # Check installation of BigDL
 if [ -z "${BIGDL_HOME}" ]; then
@@ -25,9 +24,7 @@ export BIGDL_JAR=${BIGDL_HOME}/lib/${BIGDL_JAR_NAME}
 export BIGDL_PY_ZIP_NAME=`ls ${BIGDL_HOME}/lib/ | grep python-api.zip`
 export BIGDL_PY_ZIP=${BIGDL_HOME}/lib/${BIGDL_PY_ZIP_NAME}
 export BIGDL_CONF=${BIGDL_HOME}/conf/spark-bigdl.conf
-echo ${BIGDL_JAR}
-echo ${BIGDL_PY_ZIP}
-echo ${BIGDL_CONF}
+
 # Check files
 if [ ! -f ${BIGDL_JAR} ]; then
     echo "Cannot find ${BIGDL_JAR}!"

From fff42edaeb5c81699efdfa35c461a370c7d41257 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Mon, 29 Jan 2018 16:30:39 +0800
Subject: [PATCH 03/16] Check BigDL and Spark version.

---
 toree_install.sh | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

diff --git a/toree_install.sh b/toree_install.sh
index 1db0fc5..05228b5 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -1,20 +1,37 @@
 #!/bin/bash
 
+# Required BigDL and Spark version
+export BIGDL_VERSION=0.4.0
+export SPARK_VERSION=2.2.1
+
 # Find the path of BigDL and Spark
 export BIGDL_PIP_HOME=`pip show BigDL | sed -n -e '/^Location/p' | sed 's/[^ ]* //'`
 export BIGDL_HOME=${BIGDL_PIP_HOME}/bigdl/share
 export PYSPARK_PIP_HOME=`pip show pyspark | sed -n -e '/^Location/p' | sed 's/[^ ]* //'`
-export SPARK_HOME=`python ${PYSPARK_PIP_HOME}/pyspark/find_spark_home.py`
+export SPARK_HOME=${PYSPARK_PIP_HOME}/pyspark
 
 # Check installation of BigDL
 if [ -z "${BIGDL_HOME}" ]; then
-	echo "Please install BigDL correctly!"
+	echo "Cannot find BigDL installation directory. Have you run 'pip install BigDL=${BIGDL_VERSION}'?"
 	exit 1
 fi
 
 # Check installation of Spark
 if [ -z "${SPARK_HOME}" ]; then
-	echo "Please set install Spark correctly!"
+	echo "Cannot find Spark installation directory. Have you run 'pip install BigDL=${BIGDL_VERSION}'?"
+	exit 1
+fi
+
+# Check the version BigDL and Spark
+export BIGDL_TEMP_VERSION=`pip show BigDL | sed -n -e '/^Version/p' | sed 's/[^ ]* //'`
+if [ "${BIGDL_VERSION}" != "${BIGDL_TEMP_VERSION}" ]; then
+	echo "Wrong version of BigDL. Please run 'pip install BigDL=${BIGDL_VERSION}'."
+	exit 1
+fi
+
+export SPARK_TEMP_VERSION=`pip show pyspark | sed -n -e '/^Version/p' | sed 's/[^ ]* //'`
+if [ "${SPARK_VERSION}" != "${SPARK_TEMP_VERSION}" ]; then
+        echo "Wrong version of Spark. Please run 'pip install BigDL=${BIGDL_VERSION}'."
 	exit 1
 fi
 

From 39ea406d7d33aff818ae822d396da9e2c5926834 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 11:14:37 +0800
Subject: [PATCH 04/16] Fix errors in script and add proxy configuration.

---
 toree_install.sh | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/toree_install.sh b/toree_install.sh
index 05228b5..671d8cb 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -12,26 +12,26 @@ export SPARK_HOME=${PYSPARK_PIP_HOME}/pyspark
 
 # Check installation of BigDL
 if [ -z "${BIGDL_HOME}" ]; then
-	echo "Cannot find BigDL installation directory. Have you run 'pip install BigDL=${BIGDL_VERSION}'?"
+	echo "Cannot find BigDL installation directory. Have you run 'pip install BigDL==${BIGDL_VERSION}'?"
 	exit 1
 fi
 
 # Check installation of Spark
 if [ -z "${SPARK_HOME}" ]; then
-	echo "Cannot find Spark installation directory. Have you run 'pip install BigDL=${BIGDL_VERSION}'?"
+	echo "Cannot find Spark installation directory. Have you run 'pip install BigDL==${BIGDL_VERSION}'?"
 	exit 1
 fi
 
 # Check the version BigDL and Spark
 export BIGDL_TEMP_VERSION=`pip show BigDL | sed -n -e '/^Version/p' | sed 's/[^ ]* //'`
 if [ "${BIGDL_VERSION}" != "${BIGDL_TEMP_VERSION}" ]; then
-	echo "Wrong version of BigDL. Please run 'pip install BigDL=${BIGDL_VERSION}'."
+	echo "Wrong version of BigDL. Please run 'pip install BigDL==${BIGDL_VERSION}'."
 	exit 1
 fi
 
 export SPARK_TEMP_VERSION=`pip show pyspark | sed -n -e '/^Version/p' | sed 's/[^ ]* //'`
 if [ "${SPARK_VERSION}" != "${SPARK_TEMP_VERSION}" ]; then
-        echo "Wrong version of Spark. Please run 'pip install BigDL=${BIGDL_VERSION}'."
+        echo "Wrong version of Spark. Please run 'pip install BigDL==${BIGDL_VERSION}'."
 	exit 1
 fi
 
@@ -58,8 +58,20 @@ if [ ! -f ${BIGDL_CONF} ]; then
     exit 1
 fi
 
-# Configure Spark
-export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory'"
+# Configure proxy and Spark
+if [ ! -z "{HTTP_PROXY}" ]; then
+   HTTP_PROXY_PORT=${HTTP_PROXY##*:}
+   HTTP_PROXY_NAME=${HTTP_PROXY##*/}
+   HTTP_PROXY_HOST=${HTTP_PROXY_NAME%%:*}
+fi
+
+if [ ! -z "{HTTPS_PROXY}" ]; then
+   HTTPS_PROXY_PORT=${HTTPS_PROXY##*:}
+   HTTPS_PROXY_NAME=${HTTPS_PROXY##*/}
+   HTTPS_PROXY_HOST=${HTTPS_PROXY_NAME%%:*}
+fi
+
+export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory' --driver-java-options='-Dhttp.proxyHost=${HTTP_PROXY_HOST} -Dhttp.proxyPort=${HTTP_PROXY_PORT} -Dhttps.proxyHost=${HTTPS_PROXY_HOST} -Dhttps.proxyPort=${HTTPS_PROXY_PORT}'"
 
 # Install Toree
-jupyter toree install --interpreters=Scala,PySpark --spark_home=${SPARK_HOME} --spark_opts='${SPARK_OPTS}'
+jupyter toree install --interpreters=Scala,PySpark --spark_home=${SPARK_HOME} --spark_opts="${SPARK_OPTS}"

From 015cc4db713129668e56618c01826897efe3fede Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 11:20:18 +0800
Subject: [PATCH 05/16] Refine proxy configuration.

---
 toree_install.sh | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/toree_install.sh b/toree_install.sh
index 671d8cb..e697dbc 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -59,15 +59,15 @@ if [ ! -f ${BIGDL_CONF} ]; then
 fi
 
 # Configure proxy and Spark
-if [ ! -z "{HTTP_PROXY}" ]; then
-   HTTP_PROXY_PORT=${HTTP_PROXY##*:}
-   HTTP_PROXY_NAME=${HTTP_PROXY##*/}
+if [ ! -z "{http_proxy}" ]; then
+   HTTP_PROXY_PORT=${http_proxy##*:}
+   HTTP_PROXY_NAME=${http_proxy##*/}
    HTTP_PROXY_HOST=${HTTP_PROXY_NAME%%:*}
 fi
 
-if [ ! -z "{HTTPS_PROXY}" ]; then
-   HTTPS_PROXY_PORT=${HTTPS_PROXY##*:}
-   HTTPS_PROXY_NAME=${HTTPS_PROXY##*/}
+if [ ! -z "{https_proxy}" ]; then
+   HTTPS_PROXY_PORT=${https_proxy##*:}
+   HTTPS_PROXY_NAME=${https_proxy##*/}
    HTTPS_PROXY_HOST=${HTTPS_PROXY_NAME%%:*}
 fi
 

From 1016207aa0ba2bfac5d227eb2866f7adb4b14e1b Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 11:23:47 +0800
Subject: [PATCH 06/16] Update README.md

---
 README.md | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index 71c7116..d6c213f 100644
--- a/README.md
+++ b/README.md
@@ -23,16 +23,17 @@ Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github
 + JDK 8
 + Apache Spark 2.2.0
 + Jupyter Notebook 4.1
-+ BigDL 0.3.0
++ BigDL 0.4.0
 + [Setup env on Mac OS](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupMac.md) / [Setup env on Linux](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupLinux.md)
 
-### Start Jupyter Server
+### Start Jupyter Server and Toree Kernel
 * Run ```pip install BigDL==0.3.0```
-* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser```
+* Run ```pip install https://dist.apache.org/repos/dist/dev/incubator/toree/0.2.0/snapshots/dev1/toree-pip/toree-0.2.0.dev1.tar.gz```
+* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser --NotebookApp.token='' --allow-root```
 
 ## Run Demo
 * Open a browser - Suggest Chrome or Firefox or Safari
 * Access notebook client at address http://localhost:8888, open the example ipynb files and execute.
 
 ## Note
-* This notebook is for BigDL 0.3.0. Please refer branch-0.2 if you need to use BigDL 0.2.0.
+* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0.

From 84dca49dfeac730b8b4ece8b3bbd31366532dab6 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 13:25:12 +0800
Subject: [PATCH 07/16] Refine toree_install.sh and README.

---
 README.md        | 12 ++++++------
 toree_install.sh |  8 +-------
 2 files changed, 7 insertions(+), 13 deletions(-)

diff --git a/README.md b/README.md
index d6c213f..e0124e2 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
-# Deep Leaning Tutorials on Apache Spark using BigDL
+# Deep Learning Tutorials on Apache Spark using BigDL
 
-Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
+Step-by-step Deep Learning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
 
 ### Topics
 1. [RDD](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/RDD.ipynb) 
@@ -26,14 +26,14 @@ Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github
 + BigDL 0.4.0
 + [Setup env on Mac OS](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupMac.md) / [Setup env on Linux](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupLinux.md)
 
-### Start Jupyter Server and Toree Kernel
-* Run ```pip install BigDL==0.3.0```
+### Start Jupyter Server
+* Run ```pip install BigDL==0.4.0```
 * Run ```pip install https://dist.apache.org/repos/dist/dev/incubator/toree/0.2.0/snapshots/dev1/toree-pip/toree-0.2.0.dev1.tar.gz```
-* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser --NotebookApp.token='' --allow-root```
+* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser```
 
 ## Run Demo
 * Open a browser - Suggest Chrome or Firefox or Safari
 * Access notebook client at address http://localhost:8888, open the example ipynb files and execute.
 
 ## Note
-* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0.
+* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0. ${SPARK_OPTS}
diff --git a/toree_install.sh b/toree_install.sh
index e697dbc..c0f5a8b 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -65,13 +65,7 @@ if [ ! -z "{http_proxy}" ]; then
    HTTP_PROXY_HOST=${HTTP_PROXY_NAME%%:*}
 fi
 
-if [ ! -z "{https_proxy}" ]; then
-   HTTPS_PROXY_PORT=${https_proxy##*:}
-   HTTPS_PROXY_NAME=${https_proxy##*/}
-   HTTPS_PROXY_HOST=${HTTPS_PROXY_NAME%%:*}
-fi
-
-export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory' --driver-java-options='-Dhttp.proxyHost=${HTTP_PROXY_HOST} -Dhttp.proxyPort=${HTTP_PROXY_PORT} -Dhttps.proxyHost=${HTTPS_PROXY_HOST} -Dhttps.proxyPort=${HTTPS_PROXY_PORT}'"
+export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory' --driver-java-options='-Dhttp.proxyHost=${HTTP_PROXY_HOST} -Dhttp.proxyPort=${HTTP_PROXY_PORT}'"
 
 # Install Toree
 jupyter toree install --interpreters=Scala,PySpark --spark_home=${SPARK_HOME} --spark_opts="${SPARK_OPTS}"

From f4b7d22ff3c3b1c429a1321fef16ec2b00da44bc Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 13:29:37 +0800
Subject: [PATCH 08/16] Update README.md

---
 README.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index e0124e2..180f1c7 100644
--- a/README.md
+++ b/README.md
@@ -29,11 +29,11 @@ Step-by-step Deep Learning Tutorials on Apache Spark using [BigDL](https://githu
 ### Start Jupyter Server
 * Run ```pip install BigDL==0.4.0```
 * Run ```pip install https://dist.apache.org/repos/dist/dev/incubator/toree/0.2.0/snapshots/dev1/toree-pip/toree-0.2.0.dev1.tar.gz```
-* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser```
+* Run ```jupyter notebook --notebook-dir=./ --ip=* --no-browser```
 
 ## Run Demo
 * Open a browser - Suggest Chrome or Firefox or Safari
 * Access notebook client at address http://localhost:8888, open the example ipynb files and execute.
 
 ## Note
-* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0. ${SPARK_OPTS}
+* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0.

From afea2066bc3b68455d4db941c23cf1c772a0a59c Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 13:53:15 +0800
Subject: [PATCH 09/16] Keep fork up to date.

---
 README.md | 38 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
 create mode 100644 README.md

diff --git a/README.md b/README.md
new file mode 100644
index 0000000..c42f7b1
--- /dev/null
+++ b/README.md
@@ -0,0 +1,38 @@
+# Deep Leaning Tutorials on Apache Spark using BigDL
+
+Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
+
+### Topics
+1. [RDD](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/RDD.ipynb) 
+2. [DataFrame](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/DataFrame.ipynb)
+3. [SparkSQL](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/spark_sql.ipynb)
+4. [StructureStreaming](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/structured_streaming.ipynb)
+5. [Forward and backward](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/forward_and_backward.ipynb)
+6. [Linear Regression](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/linear_regression.ipynb)
+7. [Introduction to MNIST](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/introduction_to_mnist.ipynb)
+8. [Logistic Regression](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/logistic_regression.ipynb)
+9. [Feedforward Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/deep_feed_forward_neural_network.ipynb)
+10. [Convolutional Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/cnn.ipynb)
+11. [Recurrent Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/rnn.ipynb)
+12. [LSTM](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/lstm.ipynb)
+13. [Bi-directional RNN](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/birnn.ipynb)
+14. [Auto-encoder](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/autoencoder.ipynb)
+
+### Environment
++ Python 2.7
++ JDK 8
++ Apache Spark 2.2.0
++ Jupyter Notebook 4.1
++ BigDL 0.4.0
++ [Setup env on Mac OS](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupMac.md) / [Setup env on Linux](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/SetupLinux.md)
+
+### Start Jupyter Server
+* Run ```pip install BigDL==0.4.0```
+* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser```
+
+## Run Demo
+* Open a browser - Suggest Chrome or Firefox or Safari
+* Access notebook client at address http://localhost:8888, open the example ipynb files and execute.
+
+## Note
+* This notebook is for BigDL 0.4.0. Please refer branch-0.3 if you need to use BigDL 0.3.0.

From e016752fcea7704de53084bbacaa1466f812d339 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 13:55:38 +0800
Subject: [PATCH 10/16] Delete proxy configuration.

---
 toree_install.sh | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/toree_install.sh b/toree_install.sh
index c0f5a8b..d72c205 100755
--- a/toree_install.sh
+++ b/toree_install.sh
@@ -58,14 +58,8 @@ if [ ! -f ${BIGDL_CONF} ]; then
     exit 1
 fi
 
-# Configure proxy and Spark
-if [ ! -z "{http_proxy}" ]; then
-   HTTP_PROXY_PORT=${http_proxy##*:}
-   HTTP_PROXY_NAME=${http_proxy##*/}
-   HTTP_PROXY_HOST=${HTTP_PROXY_NAME%%:*}
-fi
-
-export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory' --driver-java-options='-Dhttp.proxyHost=${HTTP_PROXY_HOST} -Dhttp.proxyPort=${HTTP_PROXY_PORT}'"
+# Configure Spark
+export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory'"
 
 # Install Toree
 jupyter toree install --interpreters=Scala,PySpark --spark_home=${SPARK_HOME} --spark_opts="${SPARK_OPTS}"

From 84cafdd5f4f46a5ec6add359edc9e54c6799c6ea Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 14:07:03 +0800
Subject: [PATCH 11/16] Update README.md

---
 README.md | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
index c42f7b1..fda2068 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
-# Deep Leaning Tutorials on Apache Spark using BigDL
+# Deep Learning Tutorials on Apache Spark using BigDL
 
-Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
+Step-by-step Deep Learning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
 
 ### Topics
 1. [RDD](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/RDD.ipynb) 
@@ -28,7 +28,13 @@ Step-by-step Deep Leaning Tutorials on Apache Spark using [BigDL](https://github
 
 ### Start Jupyter Server
 * Run ```pip install BigDL==0.4.0```
-* Run ``` jupyter notebook --notebook-dir=./ --ip=* --no-browser```
+* Run ```jupyter notebook --notebook-dir=./ --ip=* --no-browser```
+
+### Start Toree Kernel to Run Scala Notebooks
+* Run ```pip install BigDL==0.4.0```
+* Run ```pip install https://dist.apache.org/repos/dist/dev/incubator/toree/0.2.0/snapshots/dev1/toree-pip/toree-0.2.0.dev1.tar.gz```
+* Run ```./toree_install.sh```
+* Run ```jupyter notebook --notebook-dir=./ --ip=* --no-browser```
 
 ## Run Demo
 * Open a browser - Suggest Chrome or Firefox or Safari

From bbf39e753016213026b98f861f9f99e2362434ff Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 15:12:41 +0800
Subject: [PATCH 12/16] Delete start_toree.sh.

---
 start_toree.sh | 43 -------------------------------------------
 1 file changed, 43 deletions(-)
 delete mode 100755 start_toree.sh

diff --git a/start_toree.sh b/start_toree.sh
deleted file mode 100755
index 5f3c031..0000000
--- a/start_toree.sh
+++ /dev/null
@@ -1,43 +0,0 @@
-#!/bin/bash
-
-# Check environment variables
-if [ -z "${BIGDL_HOME}" ]; then
-    echo "Please set BIGDL_HOME environment variable"
-    exit 1
-fi
-
-if [ -z "${SPARK_HOME}" ]; then
-    echo "Please set SPARK_HOME environment variable"
-    exit 1
-fi
-
-#setup pathes
-export PYSPARK_DRIVER_PYTHON=jupyter
-export PYSPARK_DRIVER_PYTHON_OPTS="notebook --notebook-dir=./ --ip=* --no-browser --NotebookApp.token=''"
-export BIGDL_JAR_NAME=`ls ${BIGDL_HOME}/lib/ | grep jar-with-dependencies.jar`
-export BIGDL_JAR="${BIGDL_HOME}/lib/$BIGDL_JAR_NAME"
-export BIGDL_PY_ZIP_NAME=`ls ${BIGDL_HOME}/lib/ | grep python-api.zip`
-export BIGDL_PY_ZIP="${BIGDL_HOME}/lib/$BIGDL_PY_ZIP_NAME"
-export BIGDL_CONF=${BIGDL_HOME}/conf/spark-bigdl.conf
-
-# Check files
-if [ ! -f ${BIGDL_CONF} ]; then
-    echo "Cannot find ${BIGDL_CONF}"
-    exit 1
-fi
-
-if [ ! -f ${BIGDL_PY_ZIP} ]; then
-    echo "Cannot find ${BIGDL_PY_ZIP}"
-    exit 1
-fi
-
-if [ ! -f $BIGDL_JAR ]; then
-    echo "Cannot find $BIGDL_JAR"
-    exit 1
-fi
-
-export SPARK_OPTS="--master local[4] --driver-memory 4g --properties-file ${BIGDL_CONF} --jars ${BIGDL_JAR} --conf spark.driver.extraClassPath=${BIGDL_JAR} --conf spark.executor.extraClassPath=${BIGDL_JAR} --conf spark.sql.catalogImplementation='in-memory'"
-
-echo 'Install toree to jupyter, this may need root privilege'
-sudo jupyter toree install --spark_home=${SPARK_HOME} --spark_opts='${SPARK_OPTS}'
-jupyter notebook --notebook-dir=./ --ip=* --no-browser --NotebookApp.token=''

From 7398563f8e368a7ac1cd9c3b1e482135b8af8d1e Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 15:41:10 +0800
Subject: [PATCH 13/16] Add scala notebooks and change the structure of
 notebooks.

---
 .../neural_networks/autoencoder.ipynb         |   0
 .../{ => python}/neural_networks/birnn.ipynb  |   0
 .../{ => python}/neural_networks/cnn.ipynb    |   0
 .../deep_feed_forward_neural_network.ipynb    |   0
 .../forward_and_backward.ipynb                |   0
 .../introduction_to_mnist.ipynb               |   0
 .../neural_networks/linear_regression.ipynb   |   0
 .../neural_networks/logistic_regression.ipynb |   0
 .../{ => python}/neural_networks/lstm.ipynb   |   0
 .../{ => python}/neural_networks/rnn.ipynb    |   0
 .../Bi-directional_RNN/Bi-directional_RNN.jpg | Bin
 .../autoencoder/autoencoder_schema.jpg        | Bin
 .../feedforwardNN_structure.png               | Bin
 .../{ => python}/neural_networks/utils.py     |   0
 .../{ => python}/spark_basics/DataFrame.ipynb |   0
 notebooks/{ => python}/spark_basics/RDD.ipynb |   0
 .../{ => python}/spark_basics/spark_sql.ipynb |   0
 .../spark_basics/structured_streaming.ipynb   |   0
 .../introduction_to_mnist.ipynb               | 201 +++++++++++
 .../neural_networks/linear_regression.ipynb   | 243 +++++++++++++
 .../neural_networks/logistic_regression.ipynb | 321 ++++++++++++++++++
 21 files changed, 765 insertions(+)
 rename notebooks/{ => python}/neural_networks/autoencoder.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/birnn.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/cnn.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/deep_feed_forward_neural_network.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/forward_and_backward.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/introduction_to_mnist.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/linear_regression.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/logistic_regression.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/lstm.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/rnn.ipynb (100%)
 rename notebooks/{ => python}/neural_networks/tutorial_images/Bi-directional_RNN/Bi-directional_RNN.jpg (100%)
 rename notebooks/{ => python}/neural_networks/tutorial_images/autoencoder/autoencoder_schema.jpg (100%)
 rename notebooks/{ => python}/neural_networks/tutorial_images/deep_feed_forward_NN/feedforwardNN_structure.png (100%)
 rename notebooks/{ => python}/neural_networks/utils.py (100%)
 rename notebooks/{ => python}/spark_basics/DataFrame.ipynb (100%)
 rename notebooks/{ => python}/spark_basics/RDD.ipynb (100%)
 rename notebooks/{ => python}/spark_basics/spark_sql.ipynb (100%)
 rename notebooks/{ => python}/spark_basics/structured_streaming.ipynb (100%)
 create mode 100644 notebooks/scala/neural_networks/introduction_to_mnist.ipynb
 create mode 100644 notebooks/scala/neural_networks/linear_regression.ipynb
 create mode 100644 notebooks/scala/neural_networks/logistic_regression.ipynb

diff --git a/notebooks/neural_networks/autoencoder.ipynb b/notebooks/python/neural_networks/autoencoder.ipynb
similarity index 100%
rename from notebooks/neural_networks/autoencoder.ipynb
rename to notebooks/python/neural_networks/autoencoder.ipynb
diff --git a/notebooks/neural_networks/birnn.ipynb b/notebooks/python/neural_networks/birnn.ipynb
similarity index 100%
rename from notebooks/neural_networks/birnn.ipynb
rename to notebooks/python/neural_networks/birnn.ipynb
diff --git a/notebooks/neural_networks/cnn.ipynb b/notebooks/python/neural_networks/cnn.ipynb
similarity index 100%
rename from notebooks/neural_networks/cnn.ipynb
rename to notebooks/python/neural_networks/cnn.ipynb
diff --git a/notebooks/neural_networks/deep_feed_forward_neural_network.ipynb b/notebooks/python/neural_networks/deep_feed_forward_neural_network.ipynb
similarity index 100%
rename from notebooks/neural_networks/deep_feed_forward_neural_network.ipynb
rename to notebooks/python/neural_networks/deep_feed_forward_neural_network.ipynb
diff --git a/notebooks/neural_networks/forward_and_backward.ipynb b/notebooks/python/neural_networks/forward_and_backward.ipynb
similarity index 100%
rename from notebooks/neural_networks/forward_and_backward.ipynb
rename to notebooks/python/neural_networks/forward_and_backward.ipynb
diff --git a/notebooks/neural_networks/introduction_to_mnist.ipynb b/notebooks/python/neural_networks/introduction_to_mnist.ipynb
similarity index 100%
rename from notebooks/neural_networks/introduction_to_mnist.ipynb
rename to notebooks/python/neural_networks/introduction_to_mnist.ipynb
diff --git a/notebooks/neural_networks/linear_regression.ipynb b/notebooks/python/neural_networks/linear_regression.ipynb
similarity index 100%
rename from notebooks/neural_networks/linear_regression.ipynb
rename to notebooks/python/neural_networks/linear_regression.ipynb
diff --git a/notebooks/neural_networks/logistic_regression.ipynb b/notebooks/python/neural_networks/logistic_regression.ipynb
similarity index 100%
rename from notebooks/neural_networks/logistic_regression.ipynb
rename to notebooks/python/neural_networks/logistic_regression.ipynb
diff --git a/notebooks/neural_networks/lstm.ipynb b/notebooks/python/neural_networks/lstm.ipynb
similarity index 100%
rename from notebooks/neural_networks/lstm.ipynb
rename to notebooks/python/neural_networks/lstm.ipynb
diff --git a/notebooks/neural_networks/rnn.ipynb b/notebooks/python/neural_networks/rnn.ipynb
similarity index 100%
rename from notebooks/neural_networks/rnn.ipynb
rename to notebooks/python/neural_networks/rnn.ipynb
diff --git a/notebooks/neural_networks/tutorial_images/Bi-directional_RNN/Bi-directional_RNN.jpg b/notebooks/python/neural_networks/tutorial_images/Bi-directional_RNN/Bi-directional_RNN.jpg
similarity index 100%
rename from notebooks/neural_networks/tutorial_images/Bi-directional_RNN/Bi-directional_RNN.jpg
rename to notebooks/python/neural_networks/tutorial_images/Bi-directional_RNN/Bi-directional_RNN.jpg
diff --git a/notebooks/neural_networks/tutorial_images/autoencoder/autoencoder_schema.jpg b/notebooks/python/neural_networks/tutorial_images/autoencoder/autoencoder_schema.jpg
similarity index 100%
rename from notebooks/neural_networks/tutorial_images/autoencoder/autoencoder_schema.jpg
rename to notebooks/python/neural_networks/tutorial_images/autoencoder/autoencoder_schema.jpg
diff --git a/notebooks/neural_networks/tutorial_images/deep_feed_forward_NN/feedforwardNN_structure.png b/notebooks/python/neural_networks/tutorial_images/deep_feed_forward_NN/feedforwardNN_structure.png
similarity index 100%
rename from notebooks/neural_networks/tutorial_images/deep_feed_forward_NN/feedforwardNN_structure.png
rename to notebooks/python/neural_networks/tutorial_images/deep_feed_forward_NN/feedforwardNN_structure.png
diff --git a/notebooks/neural_networks/utils.py b/notebooks/python/neural_networks/utils.py
similarity index 100%
rename from notebooks/neural_networks/utils.py
rename to notebooks/python/neural_networks/utils.py
diff --git a/notebooks/spark_basics/DataFrame.ipynb b/notebooks/python/spark_basics/DataFrame.ipynb
similarity index 100%
rename from notebooks/spark_basics/DataFrame.ipynb
rename to notebooks/python/spark_basics/DataFrame.ipynb
diff --git a/notebooks/spark_basics/RDD.ipynb b/notebooks/python/spark_basics/RDD.ipynb
similarity index 100%
rename from notebooks/spark_basics/RDD.ipynb
rename to notebooks/python/spark_basics/RDD.ipynb
diff --git a/notebooks/spark_basics/spark_sql.ipynb b/notebooks/python/spark_basics/spark_sql.ipynb
similarity index 100%
rename from notebooks/spark_basics/spark_sql.ipynb
rename to notebooks/python/spark_basics/spark_sql.ipynb
diff --git a/notebooks/spark_basics/structured_streaming.ipynb b/notebooks/python/spark_basics/structured_streaming.ipynb
similarity index 100%
rename from notebooks/spark_basics/structured_streaming.ipynb
rename to notebooks/python/spark_basics/structured_streaming.ipynb
diff --git a/notebooks/scala/neural_networks/introduction_to_mnist.ipynb b/notebooks/scala/neural_networks/introduction_to_mnist.ipynb
new file mode 100644
index 0000000..98a2a1d
--- /dev/null
+++ b/notebooks/scala/neural_networks/introduction_to_mnist.ipynb
@@ -0,0 +1,201 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Introduction to the MNIST database"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In the following tutorials, we are going to use the MNIST database of handwritten digits. MNIST is a simple computer vision dataset of handwritten digits. It has 60,000 training examles and 10,000 test examples. \"It is a good database for people who want to try learning techniques and pattern recognition methods on real-world data while spending minimal efforts on preprocessing and formatting.\" For more details of this database, please checkout the website [MNIST](http://yann.lecun.com/exdb/mnist/)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In BigDL, we need to write a function to download and read the MNIST data when using Scala."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import java.nio.ByteBuffer\n",
+    "import java.nio.file.{Files, Path, Paths}\n",
+    "\n",
+    "import com.intel.analytics.bigdl.dataset.ByteRecord\n",
+    "import com.intel.analytics.bigdl.utils.File\n",
+    "import scopt.OptionParser\n",
+    "\n",
+    "def load(featureFile: String, labelFile: String): Array[ByteRecord] = {\n",
+    "    val featureBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(featureFile)))\n",
+    "    val labelBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(labelFile)))\n",
+    "    \n",
+    "    val labelMagicNumber = labelBuffer.getInt()\n",
+    "    require(labelMagicNumber == 2049)\n",
+    "    val featureMagicNumber = featureBuffer.getInt()\n",
+    "    require(featureMagicNumber == 2051)\n",
+    "\n",
+    "    val labelCount = labelBuffer.getInt()\n",
+    "    val featureCount = featureBuffer.getInt()\n",
+    "    require(labelCount == featureCount)\n",
+    "\n",
+    "    val rowNum = featureBuffer.getInt()\n",
+    "    val colNum = featureBuffer.getInt()\n",
+    "\n",
+    "    val result = new Array[ByteRecord](featureCount)\n",
+    "    var i = 0\n",
+    "    while (i < featureCount) {\n",
+    "      val img = new Array[Byte]((rowNum * colNum))\n",
+    "      var y = 0\n",
+    "      while (y < rowNum) {\n",
+    "        var x = 0\n",
+    "        while (x < colNum) {\n",
+    "          img(x + y * colNum) = featureBuffer.get()\n",
+    "          x += 1\n",
+    "        }\n",
+    "        y += 1\n",
+    "      }\n",
+    "      result(i) = ByteRecord(img, labelBuffer.get().toFloat + 1.0f)\n",
+    "      i += 1\n",
+    "    }\n",
+    "\n",
+    "    result\n",
+    "}"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "First, we need to import the necessary packages and initialize the engine."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import org.apache.log4j.{Level, Logger}\n",
+    "import org.apache.spark.SparkContext\n",
+    "\n",
+    "import com.intel.analytics.bigdl.utils._\n",
+    "import com.intel.analytics.bigdl.dataset.DataSet\n",
+    "import com.intel.analytics.bigdl.dataset.image.{BytesToGreyImg, GreyImgNormalizer, GreyImgToBatch, GreyImgToSample}\n",
+    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Module}\n",
+    "import com.intel.analytics.bigdl.models.lenet.Utils._\n",
+    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Linear, LogSoftMax, Sequential, Reshape}\n",
+    "import com.intel.analytics.bigdl.numeric.NumericFloat\n",
+    "import com.intel.analytics.bigdl.optim.{SGD, Top1Accuracy}\n",
+    "import com.intel.analytics.bigdl.utils.{Engine, LoggerFilter, T, Table}\n",
+    "import com.intel.analytics.bigdl.tensor.Tensor\n",
+    "\n",
+    "Engine.init"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Next, the paths of training data and validation data should be set."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "val trainData = \"../datasets/mnist/train-images-idx3-ubyte\"\n",
+    "val trainLabel = \"../datasets/mnist/train-labels-idx1-ubyte\"\n",
+    "val validationData = \"../datasets/mnist/t10k-images-idx3-ubyte\"\n",
+    "val validationLabel = \"../datasets/mnist/t10k-labels-idx1-ubyte\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "Then, we need to define some parameters for loading the MINST data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "//Parameters\n",
+    "val batchSize = 2048\n",
+    "val learningRate = 0.2\n",
+    "val maxEpochs = 15\n",
+    "\n",
+    "//Network Parameters\n",
+    "val nInput = 784 //MNIST data input (img shape: 28*28)\n",
+    "val nClasses = 10  //MNIST total classes (0-9 digits)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, we can use predefined function to load and serialize MNIST data. If you want to output the data, some modifications on the funtion should be applied."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "val trainSet = \n",
+    "    DataSet.array(load(trainData, trainLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(trainMean, trainStd) -> GreyImgToBatch(batchSize)\n",
+    "val validationSet = \n",
+    "    DataSet.array(load(validationData, validationLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(testMean, testStd) -> GreyImgToBatch(batchSize)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "sc.stop()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Apache Toree - Scala",
+   "language": "scala",
+   "name": "apache_toree_scala"
+  },
+  "language_info": {
+   "file_extension": ".scala",
+   "name": "scala",
+   "version": "2.11.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/notebooks/scala/neural_networks/linear_regression.ipynb b/notebooks/scala/neural_networks/linear_regression.ipynb
new file mode 100644
index 0000000..3f4437d
--- /dev/null
+++ b/notebooks/scala/neural_networks/linear_regression.ipynb
@@ -0,0 +1,243 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Linear Regression"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In this tutorial, we will introduce how to use BigDL to train to a simple linear regression model. The first thing we need to do it to import necessary packages and inilialize the engine."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import org.apache.log4j.{Level, Logger}\n",
+    "import org.apache.spark.SparkContext\n",
+    "\n",
+    "import com.intel.analytics.bigdl._\n",
+    "import com.intel.analytics.bigdl.utils.{Engine, LoggerFilter, T, Table}\n",
+    "import com.intel.analytics.bigdl.dataset.{DataSet, Sample}\n",
+    "import com.intel.analytics.bigdl.nn.{Sequential, Linear, MSECriterion}\n",
+    "import com.intel.analytics.bigdl.optim._\n",
+    "import com.intel.analytics.bigdl.models.lenet.Utils._\n",
+    "import com.intel.analytics.bigdl.optim.{SGD, Top1Accuracy}\n",
+    "import com.intel.analytics.bigdl.tensor._\n",
+    "import com.intel.analytics.bigdl.numeric.NumericFloat\n",
+    "\n",
+    "Engine.init"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Then we randomly create datasets for training."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "val featuresDim = 2\n",
+    "val dataLen = 100\n",
+    "\n",
+    "def GetRandSample() = {\n",
+    "    val features = Tensor(featuresDim).rand(0, 1)\n",
+    "    val label = (0.4 + features.sum * 2).toFloat\n",
+    "    val sample = Sample[Float](features, label)\n",
+    "    sample\n",
+    "}\n",
+    "\n",
+    "val rddTrain = sc.parallelize(0 until dataLen).map(_ => GetRandSample())"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "Then we specify the necessary parameters and construct a linear regression model using BigDL. Please notice that batch_size should be devided by the number of cores you use. In this example, it was set as 8 since there are 4 cores when running the example."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "// Parameters\n",
+    "val learningRate = 0.2\n",
+    "val trainingEpochs = 5\n",
+    "val batchSize = 4\n",
+    "val nInput = featuresDim\n",
+    "val nOutput = 1 \n",
+    "\n",
+    "def LinearRegression(nInput: Int, nOutput: Int) = {\n",
+    "    // Initialize a sequential container\n",
+    "    val model = Sequential()\n",
+    "    // Add a linear layer\n",
+    "    model.add(Linear(nInput, nOutput))\n",
+    "    model\n",
+    "}\n",
+    "\n",
+    "val model = LinearRegression(nInput, nOutput)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Here we construct the optimizer to optimize the linear regression problem. You can specific your own learning rate in $SGD()$ method, also, you can replace the $SGD()$ with other optimizer such like $Adam()$. Click [here](https://github.com/intel-analytics/BigDL/tree/master/spark/dl/src/main/scala/com/intel/analytics/bigdl/optim) to see more optimizer."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "com.intel.analytics.bigdl.optim.DistriOptimizer@2ec4ba5b"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "val optimizer = Optimizer(model = model, sampleRDD = rddTrain, criterion = MSECriterion[Float](), batchSize = batchSize)\n",
+    "optimizer.setOptimMethod(new SGD(learningRate=learningRate))\n",
+    "optimizer.setEndWhen(Trigger.maxEpoch(trainingEpochs))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "// Start to train\n",
+    "val trainedModel = optimizer.optimize()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Predict result:\n",
+      "3.7649865,2.7541423,1.9586959,1.5578532,3.7649865\n"
+     ]
+    }
+   ],
+   "source": [
+    "val predictResult = trainedModel.predict(rddTrain)\n",
+    "val p = predictResult.take(5).map(_.toTensor.valueAt(1)).mkString(\",\")\n",
+    "println(\"Predict result:\")\n",
+    "println(p)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "To test the trained model, we construct a dataset for testing and print the result of _Mean Square Error_."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "5.747768\n"
+     ]
+    }
+   ],
+   "source": [
+    "val r = new scala.util.Random(100)\n",
+    "val totalLength = 10\n",
+    "val features = Tensor(totalLength, featuresDim).rand(0, 1)\n",
+    "var label = (0.4 + features.sum).toFloat\n",
+    "val prediction = sc.parallelize(0 until totalLength).map(r => Sample[Float](features(r + 1), label))\n",
+    "val predictResult = trainedModel.predict(prediction)\n",
+    "val p = predictResult.take(6).map(_.toTensor.valueAt(1))\n",
+    "val groundLabel = Tensor(T(\n",
+    "                    | T(-0.47596836f),\n",
+    "                    | T(-0.37598032f),\n",
+    "                    | T(-0.00492062f),\n",
+    "                    | T(-0.5906958f),\n",
+    "                    | T(-0.12307882f),\n",
+    "                    | T(-0.77907401f)))\n",
+    "\n",
+    "var mse = 0f\n",
+    "for (i <- 1 to 6) {\n",
+    "    mse += (p(i - 1) - groundLabel(i).valueAt(1)) * (p(i - 1) - groundLabel(i).valueAt(1))\n",
+    "}\n",
+    "mse /= 6f\n",
+    "println(mse)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Finally, we stop the Spark."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "sc.stop()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Apache Toree - Scala",
+   "language": "scala",
+   "name": "apache_toree_scala"
+  },
+  "language_info": {
+   "file_extension": ".scala",
+   "name": "scala",
+   "version": "2.11.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/notebooks/scala/neural_networks/logistic_regression.ipynb b/notebooks/scala/neural_networks/logistic_regression.ipynb
new file mode 100644
index 0000000..e8a7ac4
--- /dev/null
+++ b/notebooks/scala/neural_networks/logistic_regression.ipynb
@@ -0,0 +1,321 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Logistic Regression"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In this tutorial we will introduce how to build a logistic regression model using BigDL. We use *MNIST* data for experiments in this tutorial. For more information about MNIST, please refer to this [site](http://yann.lecun.com/exdb/mnist/). The first thing we need to do it to import necessary packages and inilialize the engine."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This part aims at preparing for loading MNIST data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import java.nio.ByteBuffer\n",
+    "import java.nio.file.{Files, Path, Paths}\n",
+    "\n",
+    "import com.intel.analytics.bigdl.dataset.ByteRecord\n",
+    "import com.intel.analytics.bigdl.utils.File\n",
+    "import scopt.OptionParser\n",
+    "\n",
+    "def load(featureFile: String, labelFile: String): Array[ByteRecord] = {\n",
+    "    val featureBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(featureFile)))\n",
+    "    val labelBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(labelFile)))\n",
+    "    \n",
+    "    val labelMagicNumber = labelBuffer.getInt()\n",
+    "    require(labelMagicNumber == 2049)\n",
+    "    val featureMagicNumber = featureBuffer.getInt()\n",
+    "    require(featureMagicNumber == 2051)\n",
+    "\n",
+    "    val labelCount = labelBuffer.getInt()\n",
+    "    val featureCount = featureBuffer.getInt()\n",
+    "    require(labelCount == featureCount)\n",
+    "\n",
+    "    val rowNum = featureBuffer.getInt()\n",
+    "    val colNum = featureBuffer.getInt()\n",
+    "\n",
+    "    val result = new Array[ByteRecord](featureCount)\n",
+    "    var i = 0\n",
+    "    while (i < featureCount) {\n",
+    "      val img = new Array[Byte]((rowNum * colNum))\n",
+    "      var y = 0\n",
+    "      while (y < rowNum) {\n",
+    "        var x = 0\n",
+    "        while (x < colNum) {\n",
+    "          img(x + y * colNum) = featureBuffer.get()\n",
+    "          x += 1\n",
+    "        }\n",
+    "        y += 1\n",
+    "      }\n",
+    "      result(i) = ByteRecord(img, labelBuffer.get().toFloat + 1.0f)\n",
+    "      i += 1\n",
+    "    }\n",
+    "\n",
+    "    result\n",
+    "}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "import org.apache.log4j.{Level, Logger}\n",
+    "import org.apache.spark.SparkContext\n",
+    "\n",
+    "import com.intel.analytics.bigdl._\n",
+    "import com.intel.analytics.bigdl.utils._\n",
+    "import com.intel.analytics.bigdl.dataset.DataSet\n",
+    "import com.intel.analytics.bigdl.dataset.image.{BytesToGreyImg, GreyImgNormalizer, GreyImgToBatch, GreyImgToSample}\n",
+    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Module}\n",
+    "import com.intel.analytics.bigdl.numeric.NumericFloat\n",
+    "import com.intel.analytics.bigdl.optim._\n",
+    "import com.intel.analytics.bigdl.utils.{Engine, LoggerFilter, T, Table}\n",
+    "import com.intel.analytics.bigdl.models.lenet.Utils._\n",
+    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Linear, LogSoftMax, Sequential, Reshape}\n",
+    "import com.intel.analytics.bigdl.optim.SGD\n",
+    "import com.intel.analytics.bigdl.optim.Top1Accuracy\n",
+    "import com.intel.analytics.bigdl.tensor._"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Then we get and store MNIST for training and testing. You should edit the paths below according to your system settings."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "val trainData = \"../datasets/mnist/train-images-idx3-ubyte\"\n",
+    "val trainLabel = \"../datasets/mnist/train-labels-idx1-ubyte\"\n",
+    "val validationData = \"../datasets/mnist/t10k-images-idx3-ubyte\"\n",
+    "val validationLabel = \"../datasets/mnist/t10k-labels-idx1-ubyte\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "//Parameters\n",
+    "val batchSize = 2048\n",
+    "val learningRate = 0.2\n",
+    "val maxEpochs = 15\n",
+    "\n",
+    "//Network Parameters\n",
+    "val nInput = 784 //MNIST data input (img shape: 28*28)\n",
+    "val nClasses = 10  //MNIST total classes (0-9 digits)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "Engine.init"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {
+    "collapsed": true,
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
+    "val trainSet = \n",
+    "    DataSet.array(load(trainData, trainLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(trainMean, trainStd) -> GreyImgToBatch(batchSize)\n",
+    "val validationSet = \n",
+    "    DataSet.array(load(validationData, validationLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(testMean, testStd) -> GreyImgToBatch(batchSize)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Sequential[fca52368]{\n",
+       "  [input -> (1) -> (2) -> (3) -> output]\n",
+       "  (1): Reshape[75153ad2](784)\n",
+       "  (2): Linear[aacbecbd](784 -> 10)\n",
+       "  (3): LogSoftMax[378e7035]\n",
+       "}"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "val model = Sequential().add(Reshape(Array(28 * 28))).add(Linear(nInput, nClasses)).add(LogSoftMax())\n",
+    "model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "com.intel.analytics.bigdl.optim.DistriOptimizer@4ea3ca6e"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "val optimizer = Optimizer(model = model, dataset = trainSet, criterion = ClassNLLCriterion[Float]())\n",
+    "optimizer.setValidation(trigger = Trigger.everyEpoch, dataset = validationSet, vMethods = Array(new Top1Accuracy[Float], new Top5Accuracy[Float], new Loss[Float]))\n",
+    "optimizer.setOptimMethod(new SGD(learningRate=learningRate))\n",
+    "optimizer.setEndWhen(Trigger.maxEpoch(maxEpochs))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "can't find locality partition for partition 0 Partition locations are (ArrayBuffer(172.168.0.21)) Candidate partition locations are\n",
+      "(0,List()).\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "Sequential[fca52368]{\n",
+       "  [input -> (1) -> (2) -> (3) -> output]\n",
+       "  (1): Reshape[75153ad2](784)\n",
+       "  (2): Linear[aacbecbd](784 -> 10)\n",
+       "  (3): LogSoftMax[378e7035]\n",
+       "}"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "val trainedModel = optimizer.optimize()\n",
+    "trainedModel"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Top1Accuracy is Accuracy(correct: 9209, count: 10000, accuracy: 0.9209)\n"
+     ]
+    }
+   ],
+   "source": [
+    "val rddData = sc.parallelize(load(validationData, validationLabel), batchSize)\n",
+    "val transformer = BytesToGreyImg(28, 28) -> GreyImgNormalizer(testMean, testStd) -> GreyImgToSample()\n",
+    "val evaluationSet = transformer(rddData)\n",
+    "        \n",
+    "val result = model.evaluate(evaluationSet, Array(new Top1Accuracy[Float]), Some(batchSize))\n",
+    "\n",
+    "result.foreach(r => println(s\"${r._2} is ${r._1}\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "8.0,3.0,2.0,1.0,5.0,2.0,5.0,10.0,7.0,10.0,1.0,7.0,10.0,1.0,2.0,6.0,10.0,8.0,4.0,5.0\n",
+      "8.0,3.0,2.0,1.0,5.0,2.0,5.0,10.0,6.0,10.0,1.0,7.0,10.0,1.0,2.0,6.0,10.0,8.0,4.0,5.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "val predictions = model.predict(evaluationSet)\n",
+    "val preLabels = predictions.take(20).map(_.toTensor.max(1)._2.valueAt(1)).mkString(\",\")\n",
+    "val labels = evaluationSet.take(20).map(_.label.valueAt(1)).mkString(\",\")\n",
+    "println(preLabels)\n",
+    "println(labels)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "sc.stop()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Apache Toree - Scala",
+   "language": "scala",
+   "name": "apache_toree_scala"
+  },
+  "language_info": {
+   "file_extension": ".scala",
+   "name": "scala",
+   "version": "2.11.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

From 2f9c3e5c53dd42ed2a78c709f7877c0a5907fc10 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 15:52:16 +0800
Subject: [PATCH 14/16] Modify hyperlinks in README.md.

---
 README.md | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/README.md b/README.md
index fda2068..85a745c 100644
--- a/README.md
+++ b/README.md
@@ -3,20 +3,20 @@
 Step-by-step Deep Learning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
 
 ### Topics
-1. [RDD](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/RDD.ipynb) 
-2. [DataFrame](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/DataFrame.ipynb)
-3. [SparkSQL](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/spark_sql.ipynb)
-4. [StructureStreaming](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/spark_basics/structured_streaming.ipynb)
-5. [Forward and backward](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/forward_and_backward.ipynb)
-6. [Linear Regression](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/linear_regression.ipynb)
-7. [Introduction to MNIST](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/introduction_to_mnist.ipynb)
-8. [Logistic Regression](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/logistic_regression.ipynb)
-9. [Feedforward Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/deep_feed_forward_neural_network.ipynb)
-10. [Convolutional Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/cnn.ipynb)
-11. [Recurrent Neural Network](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/rnn.ipynb)
-12. [LSTM](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/lstm.ipynb)
-13. [Bi-directional RNN](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/birnn.ipynb)
-14. [Auto-encoder](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/neural_networks/autoencoder.ipynb)
+1. RDD [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/RDD.ipynb) 
+2. DataFrame [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/DataFrame.ipynb)
+3. SparkSQL [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/spark_sql.ipynb)
+4. StructureStreaming [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/structured_streaming.ipynb)
+5. Forward and backward [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/forward_and_backward.ipynb)
+6. Linear Regression [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/linear_regression.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/linear_regression.ipynb)
+7. Introduction to MNIST [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/introduction_to_mnist.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/introduction_to_mnist.ipynb)
+8. Logistic Regression [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/logistic_regression.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/logistic_regression.ipynb)
+9. Feedforward Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/deep_feed_forward_neural_network.ipynb)
+10. Convolutional Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/cnn.ipynb)
+11. Recurrent Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/rnn.ipynb)
+12. LSTM [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/lstm.ipynb)
+13. Bi-directional RNN[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/birnn.ipynb)
+14. Auto-encoder [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/autoencoder.ipynb)
 
 ### Environment
 + Python 2.7

From d4256e60e9aba3cfdac1a155a0ec1a7887877185 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Thu, 1 Feb 2018 16:36:35 +0800
Subject: [PATCH 15/16] Update README.md

---
 README.md | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/README.md b/README.md
index 85a745c..eea1673 100644
--- a/README.md
+++ b/README.md
@@ -3,20 +3,20 @@
 Step-by-step Deep Learning Tutorials on Apache Spark using [BigDL](https://github.com/intel-analytics/BigDL/). The tutorials are inspired by [Apache Spark examples](http://spark.apache.org/examples.html), the [Theano Tutorials](https://github.com/Newmu/Theano-Tutorials) and the [Tensorflow tutorials](https://github.com/nlintz/TensorFlow-Tutorials).
 
 ### Topics
-1. RDD [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/RDD.ipynb) 
-2. DataFrame [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/DataFrame.ipynb)
-3. SparkSQL [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/spark_sql.ipynb)
-4. StructureStreaming [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/structured_streaming.ipynb)
-5. Forward and backward [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/forward_and_backward.ipynb)
-6. Linear Regression [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/linear_regression.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/linear_regression.ipynb)
-7. Introduction to MNIST [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/introduction_to_mnist.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/introduction_to_mnist.ipynb)
-8. Logistic Regression [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/logistic_regression.ipynb) [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/logistic_regression.ipynb)
-9. Feedforward Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/deep_feed_forward_neural_network.ipynb)
-10. Convolutional Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/cnn.ipynb)
-11. Recurrent Neural Network [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/rnn.ipynb)
-12. LSTM [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/lstm.ipynb)
-13. Bi-directional RNN[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/birnn.ipynb)
-14. Auto-encoder [Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/autoencoder.ipynb)
+1. RDD [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/RDD.ipynb)]
+2. DataFrame [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/DataFrame.ipynb)]
+3. SparkSQL [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/spark_sql.ipynb)]
+4. StructureStreaming [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/spark_basics/structured_streaming.ipynb)]
+5. Forward and backward [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/forward_and_backward.ipynb)]
+6. Linear Regression [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/linear_regression.ipynb) | [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/linear_regression.ipynb)]
+7. Introduction to MNIST [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/introduction_to_mnist.ipynb) | [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/introduction_to_mnist.ipynb)]
+8. Logistic Regression [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/logistic_regression.ipynb) | [Scala](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/scala/neural_networks/logistic_regression.ipynb)]
+9. Feedforward Neural Network [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/deep_feed_forward_neural_network.ipynb)]
+10. Convolutional Neural Network [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/cnn.ipynb)]
+11. Recurrent Neural Network [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/rnn.ipynb)]
+12. LSTM [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/lstm.ipynb)]
+13. Bi-directional RNN [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/birnn.ipynb)]
+14. Auto-encoder [[Python](https://github.com/intel-analytics/BigDL-Tutorials/blob/master/notebooks/python/neural_networks/autoencoder.ipynb)]
 
 ### Environment
 + Python 2.7

From 55ec57adfc2c6b6a516cc3ce2f604c1edab65dc1 Mon Sep 17 00:00:00 2001
From: Calculuser <alvin_zuo@sjtu.edu.cn>
Date: Fri, 2 Feb 2018 13:35:45 +0800
Subject: [PATCH 16/16] Delete linear regression and logistic regression.

---
 .../neural_networks/linear_regression.ipynb   | 243 -------------
 .../neural_networks/logistic_regression.ipynb | 321 ------------------
 2 files changed, 564 deletions(-)
 delete mode 100644 notebooks/scala/neural_networks/linear_regression.ipynb
 delete mode 100644 notebooks/scala/neural_networks/logistic_regression.ipynb

diff --git a/notebooks/scala/neural_networks/linear_regression.ipynb b/notebooks/scala/neural_networks/linear_regression.ipynb
deleted file mode 100644
index 3f4437d..0000000
--- a/notebooks/scala/neural_networks/linear_regression.ipynb
+++ /dev/null
@@ -1,243 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Linear Regression"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "In this tutorial, we will introduce how to use BigDL to train to a simple linear regression model. The first thing we need to do it to import necessary packages and inilialize the engine."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "import org.apache.log4j.{Level, Logger}\n",
-    "import org.apache.spark.SparkContext\n",
-    "\n",
-    "import com.intel.analytics.bigdl._\n",
-    "import com.intel.analytics.bigdl.utils.{Engine, LoggerFilter, T, Table}\n",
-    "import com.intel.analytics.bigdl.dataset.{DataSet, Sample}\n",
-    "import com.intel.analytics.bigdl.nn.{Sequential, Linear, MSECriterion}\n",
-    "import com.intel.analytics.bigdl.optim._\n",
-    "import com.intel.analytics.bigdl.models.lenet.Utils._\n",
-    "import com.intel.analytics.bigdl.optim.{SGD, Top1Accuracy}\n",
-    "import com.intel.analytics.bigdl.tensor._\n",
-    "import com.intel.analytics.bigdl.numeric.NumericFloat\n",
-    "\n",
-    "Engine.init"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Then we randomly create datasets for training."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "val featuresDim = 2\n",
-    "val dataLen = 100\n",
-    "\n",
-    "def GetRandSample() = {\n",
-    "    val features = Tensor(featuresDim).rand(0, 1)\n",
-    "    val label = (0.4 + features.sum * 2).toFloat\n",
-    "    val sample = Sample[Float](features, label)\n",
-    "    sample\n",
-    "}\n",
-    "\n",
-    "val rddTrain = sc.parallelize(0 until dataLen).map(_ => GetRandSample())"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {
-    "collapsed": true
-   },
-   "source": [
-    "Then we specify the necessary parameters and construct a linear regression model using BigDL. Please notice that batch_size should be devided by the number of cores you use. In this example, it was set as 8 since there are 4 cores when running the example."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "// Parameters\n",
-    "val learningRate = 0.2\n",
-    "val trainingEpochs = 5\n",
-    "val batchSize = 4\n",
-    "val nInput = featuresDim\n",
-    "val nOutput = 1 \n",
-    "\n",
-    "def LinearRegression(nInput: Int, nOutput: Int) = {\n",
-    "    // Initialize a sequential container\n",
-    "    val model = Sequential()\n",
-    "    // Add a linear layer\n",
-    "    model.add(Linear(nInput, nOutput))\n",
-    "    model\n",
-    "}\n",
-    "\n",
-    "val model = LinearRegression(nInput, nOutput)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Here we construct the optimizer to optimize the linear regression problem. You can specific your own learning rate in $SGD()$ method, also, you can replace the $SGD()$ with other optimizer such like $Adam()$. Click [here](https://github.com/intel-analytics/BigDL/tree/master/spark/dl/src/main/scala/com/intel/analytics/bigdl/optim) to see more optimizer."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "com.intel.analytics.bigdl.optim.DistriOptimizer@2ec4ba5b"
-      ]
-     },
-     "execution_count": 4,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "val optimizer = Optimizer(model = model, sampleRDD = rddTrain, criterion = MSECriterion[Float](), batchSize = batchSize)\n",
-    "optimizer.setOptimMethod(new SGD(learningRate=learningRate))\n",
-    "optimizer.setEndWhen(Trigger.maxEpoch(trainingEpochs))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "// Start to train\n",
-    "val trainedModel = optimizer.optimize()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Predict result:\n",
-      "3.7649865,2.7541423,1.9586959,1.5578532,3.7649865\n"
-     ]
-    }
-   ],
-   "source": [
-    "val predictResult = trainedModel.predict(rddTrain)\n",
-    "val p = predictResult.take(5).map(_.toTensor.valueAt(1)).mkString(\",\")\n",
-    "println(\"Predict result:\")\n",
-    "println(p)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "To test the trained model, we construct a dataset for testing and print the result of _Mean Square Error_."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "5.747768\n"
-     ]
-    }
-   ],
-   "source": [
-    "val r = new scala.util.Random(100)\n",
-    "val totalLength = 10\n",
-    "val features = Tensor(totalLength, featuresDim).rand(0, 1)\n",
-    "var label = (0.4 + features.sum).toFloat\n",
-    "val prediction = sc.parallelize(0 until totalLength).map(r => Sample[Float](features(r + 1), label))\n",
-    "val predictResult = trainedModel.predict(prediction)\n",
-    "val p = predictResult.take(6).map(_.toTensor.valueAt(1))\n",
-    "val groundLabel = Tensor(T(\n",
-    "                    | T(-0.47596836f),\n",
-    "                    | T(-0.37598032f),\n",
-    "                    | T(-0.00492062f),\n",
-    "                    | T(-0.5906958f),\n",
-    "                    | T(-0.12307882f),\n",
-    "                    | T(-0.77907401f)))\n",
-    "\n",
-    "var mse = 0f\n",
-    "for (i <- 1 to 6) {\n",
-    "    mse += (p(i - 1) - groundLabel(i).valueAt(1)) * (p(i - 1) - groundLabel(i).valueAt(1))\n",
-    "}\n",
-    "mse /= 6f\n",
-    "println(mse)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Finally, we stop the Spark."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "sc.stop()"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Apache Toree - Scala",
-   "language": "scala",
-   "name": "apache_toree_scala"
-  },
-  "language_info": {
-   "file_extension": ".scala",
-   "name": "scala",
-   "version": "2.11.8"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}
diff --git a/notebooks/scala/neural_networks/logistic_regression.ipynb b/notebooks/scala/neural_networks/logistic_regression.ipynb
deleted file mode 100644
index e8a7ac4..0000000
--- a/notebooks/scala/neural_networks/logistic_regression.ipynb
+++ /dev/null
@@ -1,321 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Logistic Regression"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "In this tutorial we will introduce how to build a logistic regression model using BigDL. We use *MNIST* data for experiments in this tutorial. For more information about MNIST, please refer to this [site](http://yann.lecun.com/exdb/mnist/). The first thing we need to do it to import necessary packages and inilialize the engine."
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This part aims at preparing for loading MNIST data"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "import java.nio.ByteBuffer\n",
-    "import java.nio.file.{Files, Path, Paths}\n",
-    "\n",
-    "import com.intel.analytics.bigdl.dataset.ByteRecord\n",
-    "import com.intel.analytics.bigdl.utils.File\n",
-    "import scopt.OptionParser\n",
-    "\n",
-    "def load(featureFile: String, labelFile: String): Array[ByteRecord] = {\n",
-    "    val featureBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(featureFile)))\n",
-    "    val labelBuffer = ByteBuffer.wrap(Files.readAllBytes(Paths.get(labelFile)))\n",
-    "    \n",
-    "    val labelMagicNumber = labelBuffer.getInt()\n",
-    "    require(labelMagicNumber == 2049)\n",
-    "    val featureMagicNumber = featureBuffer.getInt()\n",
-    "    require(featureMagicNumber == 2051)\n",
-    "\n",
-    "    val labelCount = labelBuffer.getInt()\n",
-    "    val featureCount = featureBuffer.getInt()\n",
-    "    require(labelCount == featureCount)\n",
-    "\n",
-    "    val rowNum = featureBuffer.getInt()\n",
-    "    val colNum = featureBuffer.getInt()\n",
-    "\n",
-    "    val result = new Array[ByteRecord](featureCount)\n",
-    "    var i = 0\n",
-    "    while (i < featureCount) {\n",
-    "      val img = new Array[Byte]((rowNum * colNum))\n",
-    "      var y = 0\n",
-    "      while (y < rowNum) {\n",
-    "        var x = 0\n",
-    "        while (x < colNum) {\n",
-    "          img(x + y * colNum) = featureBuffer.get()\n",
-    "          x += 1\n",
-    "        }\n",
-    "        y += 1\n",
-    "      }\n",
-    "      result(i) = ByteRecord(img, labelBuffer.get().toFloat + 1.0f)\n",
-    "      i += 1\n",
-    "    }\n",
-    "\n",
-    "    result\n",
-    "}"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "import org.apache.log4j.{Level, Logger}\n",
-    "import org.apache.spark.SparkContext\n",
-    "\n",
-    "import com.intel.analytics.bigdl._\n",
-    "import com.intel.analytics.bigdl.utils._\n",
-    "import com.intel.analytics.bigdl.dataset.DataSet\n",
-    "import com.intel.analytics.bigdl.dataset.image.{BytesToGreyImg, GreyImgNormalizer, GreyImgToBatch, GreyImgToSample}\n",
-    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Module}\n",
-    "import com.intel.analytics.bigdl.numeric.NumericFloat\n",
-    "import com.intel.analytics.bigdl.optim._\n",
-    "import com.intel.analytics.bigdl.utils.{Engine, LoggerFilter, T, Table}\n",
-    "import com.intel.analytics.bigdl.models.lenet.Utils._\n",
-    "import com.intel.analytics.bigdl.nn.{ClassNLLCriterion, Linear, LogSoftMax, Sequential, Reshape}\n",
-    "import com.intel.analytics.bigdl.optim.SGD\n",
-    "import com.intel.analytics.bigdl.optim.Top1Accuracy\n",
-    "import com.intel.analytics.bigdl.tensor._"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Then we get and store MNIST for training and testing. You should edit the paths below according to your system settings."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "val trainData = \"../datasets/mnist/train-images-idx3-ubyte\"\n",
-    "val trainLabel = \"../datasets/mnist/train-labels-idx1-ubyte\"\n",
-    "val validationData = \"../datasets/mnist/t10k-images-idx3-ubyte\"\n",
-    "val validationLabel = \"../datasets/mnist/t10k-labels-idx1-ubyte\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "//Parameters\n",
-    "val batchSize = 2048\n",
-    "val learningRate = 0.2\n",
-    "val maxEpochs = 15\n",
-    "\n",
-    "//Network Parameters\n",
-    "val nInput = 784 //MNIST data input (img shape: 28*28)\n",
-    "val nClasses = 10  //MNIST total classes (0-9 digits)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "Engine.init"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {
-    "collapsed": true,
-    "scrolled": true
-   },
-   "outputs": [],
-   "source": [
-    "val trainSet = \n",
-    "    DataSet.array(load(trainData, trainLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(trainMean, trainStd) -> GreyImgToBatch(batchSize)\n",
-    "val validationSet = \n",
-    "    DataSet.array(load(validationData, validationLabel), sc) -> BytesToGreyImg(28, 28) -> GreyImgNormalizer(testMean, testStd) -> GreyImgToBatch(batchSize)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Sequential[fca52368]{\n",
-       "  [input -> (1) -> (2) -> (3) -> output]\n",
-       "  (1): Reshape[75153ad2](784)\n",
-       "  (2): Linear[aacbecbd](784 -> 10)\n",
-       "  (3): LogSoftMax[378e7035]\n",
-       "}"
-      ]
-     },
-     "execution_count": 7,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "val model = Sequential().add(Reshape(Array(28 * 28))).add(Linear(nInput, nClasses)).add(LogSoftMax())\n",
-    "model"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "com.intel.analytics.bigdl.optim.DistriOptimizer@4ea3ca6e"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "val optimizer = Optimizer(model = model, dataset = trainSet, criterion = ClassNLLCriterion[Float]())\n",
-    "optimizer.setValidation(trigger = Trigger.everyEpoch, dataset = validationSet, vMethods = Array(new Top1Accuracy[Float], new Top5Accuracy[Float], new Loss[Float]))\n",
-    "optimizer.setOptimMethod(new SGD(learningRate=learningRate))\n",
-    "optimizer.setEndWhen(Trigger.maxEpoch(maxEpochs))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "can't find locality partition for partition 0 Partition locations are (ArrayBuffer(172.168.0.21)) Candidate partition locations are\n",
-      "(0,List()).\n"
-     ]
-    },
-    {
-     "data": {
-      "text/plain": [
-       "Sequential[fca52368]{\n",
-       "  [input -> (1) -> (2) -> (3) -> output]\n",
-       "  (1): Reshape[75153ad2](784)\n",
-       "  (2): Linear[aacbecbd](784 -> 10)\n",
-       "  (3): LogSoftMax[378e7035]\n",
-       "}"
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "val trainedModel = optimizer.optimize()\n",
-    "trainedModel"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Top1Accuracy is Accuracy(correct: 9209, count: 10000, accuracy: 0.9209)\n"
-     ]
-    }
-   ],
-   "source": [
-    "val rddData = sc.parallelize(load(validationData, validationLabel), batchSize)\n",
-    "val transformer = BytesToGreyImg(28, 28) -> GreyImgNormalizer(testMean, testStd) -> GreyImgToSample()\n",
-    "val evaluationSet = transformer(rddData)\n",
-    "        \n",
-    "val result = model.evaluate(evaluationSet, Array(new Top1Accuracy[Float]), Some(batchSize))\n",
-    "\n",
-    "result.foreach(r => println(s\"${r._2} is ${r._1}\"))"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "8.0,3.0,2.0,1.0,5.0,2.0,5.0,10.0,7.0,10.0,1.0,7.0,10.0,1.0,2.0,6.0,10.0,8.0,4.0,5.0\n",
-      "8.0,3.0,2.0,1.0,5.0,2.0,5.0,10.0,6.0,10.0,1.0,7.0,10.0,1.0,2.0,6.0,10.0,8.0,4.0,5.0\n"
-     ]
-    }
-   ],
-   "source": [
-    "val predictions = model.predict(evaluationSet)\n",
-    "val preLabels = predictions.take(20).map(_.toTensor.max(1)._2.valueAt(1)).mkString(\",\")\n",
-    "val labels = evaluationSet.take(20).map(_.label.valueAt(1)).mkString(\",\")\n",
-    "println(preLabels)\n",
-    "println(labels)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {
-    "collapsed": true
-   },
-   "outputs": [],
-   "source": [
-    "sc.stop()"
-   ]
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Apache Toree - Scala",
-   "language": "scala",
-   "name": "apache_toree_scala"
-  },
-  "language_info": {
-   "file_extension": ".scala",
-   "name": "scala",
-   "version": "2.11.8"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}