diff --git a/.travis.yml b/.travis.yml
index 387367a2305e7bf582e29538ab9e51571b9ae75b..44b755ee32d204c883f0d74e7ad0f78380918954 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -49,6 +49,7 @@ before_install:
   # Paddle is using protobuf 3.1 currently. Protobuf 3.2 breaks the compatibility. So we specify the python 
   # protobuf version.
   - pip install numpy wheel 'protobuf==3.1' sphinx==1.5.6 recommonmark sphinx-rtd-theme==0.1.9 virtualenv pre-commit requests==2.9.2 LinkChecker
+  - pip install rarfile
   - |
     function timeout() { perl -e 'alarm shift; exec @ARGV' "$@"; }
 script:
diff --git a/Dockerfile b/Dockerfile
index 571c3e1476e1f3cb0750b6fffa74ee1625a7d19d..b6f99ca539d077164c71d797a5ccda7b1b5c44ba 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -56,7 +56,8 @@ RUN pip install --upgrade pip && \
     pip install -U docopt PyYAML sphinx && \
     pip install -U sphinx-rtd-theme==0.1.9 recommonmark && \
     pip install pre-commit 'requests==2.9.2' 'ipython==5.3.0' && \
-    pip install 'ipykernel==4.6.0' 'jupyter==1.0.0'
+    pip install 'ipykernel==4.6.0' 'jupyter==1.0.0' && \ 
+    pip install rarfile
 
 # To fix https://github.com/PaddlePaddle/Paddle/issues/1954, we use
 # the solution in https://urllib3.readthedocs.io/en/latest/user-guide.html#ssl-py2
diff --git a/paddle/scripts/run_python_tests.sh b/paddle/scripts/run_python_tests.sh
index 02d2cdb977473c1032b06ffca59544b3ba98d1fa..c588b9e08def5510e5daea2fd4e18f66c5f89bb8 100755
--- a/paddle/scripts/run_python_tests.sh
+++ b/paddle/scripts/run_python_tests.sh
@@ -29,7 +29,7 @@ if [ $USE_VIRTUALENV_FOR_TEST -ne 0 ]; then
 fi
 
 export PYTHONPATH=$SCRIPTPATH/../../python/
-$PYTHON -m pip install $SCRIPTPATH/../dist/*.whl requests matplotlib opencv-python ipython==5.3
+$PYTHON -m pip install $SCRIPTPATH/../dist/*.whl requests matplotlib opencv-python ipython==5.3 rarfile
 
 for fn in "$@"
 do
diff --git a/python/paddle/v2/dataset/__init__.py b/python/paddle/v2/dataset/__init__.py
index 80ff6295c34e853d8f69b9e78719af23a56d1fbb..26252d5bbd77ddb70b4f03843679e4737e2f96d3 100644
--- a/python/paddle/v2/dataset/__init__.py
+++ b/python/paddle/v2/dataset/__init__.py
@@ -24,8 +24,9 @@ import conll05
 import uci_housing
 import sentiment
 import wmt14
+import mq2007
 
 __all__ = [
     'mnist', 'imikolov', 'imdb', 'cifar', 'movielens', 'conll05', 'sentiment'
-    'uci_housing', 'wmt14'
+    'uci_housing', 'wmt14', 'mq2007'
 ]
diff --git a/python/setup.py.in b/python/setup.py.in
index 5dfb46192ae54fdc36b0867312cf156aefb84f84..1afaffd261702738cdd552636a26dfcfbfe5ac48 100644
--- a/python/setup.py.in
+++ b/python/setup.py.in
@@ -1,5 +1,6 @@
 from setuptools import setup
 
+
 packages=['paddle',
           'paddle.proto',
           'paddle.trainer',
@@ -18,6 +19,7 @@ setup(name='paddle',
           "numpy",
           "protobuf==${PROTOBUF_VERSION}",
           "matplotlib",
+          "rarfile"
       ],
       packages=packages,
       package_dir={