Restore and Update Dockerfile (#5)

.dockerignore

@@ -1,2 +1 @@
-build/
-tests/test_resources
+build/

Dockerfile

@@ -0,0 +1,46 @@
+FROM python:3.6-slim-stretch
+
+RUN mkdir -p /python-deps/Willump
+
+RUN apt-get update
+RUN apt-get install -y build-essential curl
+RUN apt-get update
+
+RUN curl https://sh.rustup.rs -sSf | bash -s -- -y
+RUN echo 'source $HOME/.cargo/env' >> $HOME/.bashrc
+
+RUN apt-get install -y zlib1g-dev wget software-properties-common
+RUN wget https://apt.llvm.org/llvm-snapshot.gpg.key -O bob.key
+RUN apt-key add bob.key
+RUN apt-add-repository "deb http://apt.llvm.org/stretch/ llvm-toolchain-stretch-6.0 main"
+RUN apt-get update
+RUN apt-get install -y llvm-6.0-dev clang-6.0
+RUN ln -s /usr/bin/llvm-config-6.0 /usr/local/bin/llvm-config
+RUN ln -s /usr/bin/clang++-6.0 /usr/local/bin/clang++
+
+RUN apt-get install -y gcc git
+ENV PATH="/root/.cargo/bin:${PATH}"
+RUN git clone https://github.com/stanford-futuredata/weld-willump.git
+WORKDIR "/weld-willump"
+RUN git checkout llvm-st
+RUN cargo build --release
+
+ENV WELD_HOME=/weld-willump
+RUN pip install setuptools
+WORKDIR "/weld-willump/python/pyweld"
+RUN python setup.py install
+WORKDIR "/"
+RUN cp weld-willump/target/release/libweld.so /usr/lib/libweld.so
+
+ADD $WILLUMP_HOME/tests /python-deps/Willump/tests
+ADD $WILLUMP_HOME/willump /python-deps/Willump/willump
+ADD $WILLUMP_HOME/cppextensions /python-deps/Willump/cppextensions
+ADD $WILLUMP_HOME/requirements.txt /python-deps/Willump
+ADD $WILLUMP_HOME/setup.py /python-deps/Willump
+ADD $WILLUMP_HOME/simple_benchmarks.sh /python-deps/Willump
+
+ENV WILLUMP_HOME=/python-deps/Willump
+
+WORKDIR "/python-deps/Willump"
+RUN pip install -r requirements.txt
+RUN python setup.py install --user

README.md

@@ -44,4 +44,18 @@ To confirm Willump works, run the Willump unit tests:
     python3 setup.py test
 
 For information on reproducing the experiments run in the Willump paper, please see our
-[benchmarks guide](https://github.com/stanford-futuredata/Willump/blob/master/BENCHMARKS.md).
+[benchmarks guide](https://github.com/stanford-futuredata/Willump/blob/master/BENCHMARKS.md).
+
+## Docker
+
+You can also experiment with Willump using Docker with the Dockerfile we provide.  To clone our repository
+and build our container, run:
+
+    git clone https://github.com/stanford-futuredata/Willump.git
+    cd Willump
+    export WILLUMP_HOME=`pwd`
+    docker build -t willump .
+
+Then, to verify the container built successfully, run our unit tests:
+
+    docker run -t willump python setup.py test

tests/benchmark_scripts/music_point.py

@@ -170,7 +170,7 @@ def create_featureset(folder):
 
     # add latent features
     y_pred = np.hstack(add_features_and_predict(folder, combi))
-    print("Train AUC: %f" % willump_score_function(y, y_pred))
+    print("Validation AUC: %f" % willump_score_function(y, y_pred))
 
 
 if __name__ == '__main__':