[DO NOT MERGE] Preliminary Ubuntu 20.04 images for both CPU and GPU as well as docs

docker/GenCPU.dockerfile

@@ -0,0 +1,40 @@
+FROM ubuntu:20.04
+
+LABEL maintainer="MIT Probabilistic Computing Project"
+
+# Find current Julia version on https://julialang.org/downloads/
+ARG JULIA_VERSION_SHORT="1.5"
+ARG JULIA_VERSION_FULL="${JULIA_VERSION_SHORT}.1"
+ENV JULIA_INSTALLATION_PATH=/opt/julia
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV JULIA_INSTALLATION_PATH=/opt/julia
+
+RUN apt-get update -qq \
+    && apt-get install -qq -y --no-install-recommends\
+        build-essential \
+        ca-certificates \
+        curl \
+        ffmpeg \
+        git \
+        graphviz \
+        hdf5-tools \
+        python3-dev \
+        python3-pip \
+        python3-tk \
+        rsync \
+        software-properties-common \
+        wget \
+        zlib1g-dev \
+    && rm -rf /var/lib/apt/lists/* && \
+    ln -s /usr/bin/python3 /usr/bin/python
+
+RUN wget https://julialang-s3.julialang.org/bin/linux/x64/${JULIA_VERSION_SHORT}/julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    tar zxf julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    mkdir -p "${JULIA_INSTALLATION_PATH}" && \
+    mv julia-${JULIA_VERSION_FULL} "${JULIA_INSTALLATION_PATH}/" && \
+    ln -fs "${JULIA_INSTALLATION_PATH}/julia-${JULIA_VERSION_FULL}/bin/julia" /usr/local/bin/ && \
+    rm julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    julia -e 'import Pkg; Pkg.add("IJulia")'
+
+RUN julia -e 'import Pkg; Pkg.add(["Gen"])'

docker/GenGPU.dockerfile

@@ -0,0 +1,40 @@
+FROM nvidia/opengl:1.0-glvnd-runtime-ubuntu20.04
+
+LABEL maintainer="MIT Probabilistic Computing Project"
+
+# Find current Julia version on https://julialang.org/downloads/
+ARG JULIA_VERSION_SHORT="1.5"
+ARG JULIA_VERSION_FULL="${JULIA_VERSION_SHORT}.1"
+ENV JULIA_INSTALLATION_PATH=/opt/julia
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV JULIA_INSTALLATION_PATH=/opt/julia
+
+RUN apt-get update -qq \
+    && apt-get install -qq -y --no-install-recommends\
+        build-essential \
+        ca-certificates \
+        curl \
+        ffmpeg \
+        git \
+        graphviz \
+        hdf5-tools \
+        python3-dev \
+        python3-pip \
+        python3-tk \
+        rsync \
+        software-properties-common \
+        wget \
+        zlib1g-dev \
+    && rm -rf /var/lib/apt/lists/* && \
+    ln -s /usr/bin/python3 /usr/bin/python
+
+RUN wget https://julialang-s3.julialang.org/bin/linux/x64/${JULIA_VERSION_SHORT}/julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    tar zxf julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    mkdir -p "${JULIA_INSTALLATION_PATH}" && \
+    mv julia-${JULIA_VERSION_FULL} "${JULIA_INSTALLATION_PATH}/" && \
+    ln -fs "${JULIA_INSTALLATION_PATH}/julia-${JULIA_VERSION_FULL}/bin/julia" /usr/local/bin/ && \
+    rm julia-${JULIA_VERSION_FULL}-linux-x86_64.tar.gz && \
+    julia -e 'import Pkg; Pkg.add("IJulia")'
+
+RUN julia -e 'import Pkg; Pkg.add(["Gen"])'

docker/Readme.md

@@ -0,0 +1,41 @@
+# Gen Containerization
+
+## Quickstart
+
+### Docker Setup
+Please follow the [official Docker installation instructions](https://docs.docker.com/engine/install/) to setup Docker on your target platform. If you would like to use GPU acceleration, please also install the [NVIDIA Container Toolkit / nvidia-docker](https://github.com/NVIDIA/nvidia-docker).
+
+
+### Building Container
+You can build Docker containers for Gen that either target CPU-only or GPU-enabled machines.
+- Please run `docker build -f docker/GenCPU.dockerfile -t gen:cpu-ubuntu20.04 .` for the CPU image
+- Please run `docker build -f docker/GenGPU.dockerfile -t gen:gpu-ubuntu20.04 .` if you want to use GPU acceleration
+Both commands are supposed to be run from the main Gen.jl folder, not this subfolder.
+
+### Running Container
+You can run `docker run -it gen:cpu-ubuntu20.04 bash` to run the CPU image and `docker run -it --gpus all gen:gpu-ubuntu20.04 bash` to run the GPU image. Afterwards, you can just execute `julia`, type `using Gen` and start developing with Gen. If you would like to remove the container, please add the flag `--rm` to your run command.
+
+## Known Issues
+
+### Run or Build Script Not Executable
+If you get an error like
+```
+bash: ./docker_build_gpu.sh: Permission denied
+```
+Then please make the script executable via `chmod +x docker_build_gpu.sh` (analogous for CPU).
+
+### Known Issues with Docker on macOS
+The Docker container might currently fail on macOS with the following error:
+```
+julia> using Gen
+[ Info: Precompiling Gen [ea4f424c-a589-11e8-07c0-fd5c91b9da4a]
+ERROR: Failed to precompile Gen [ea4f424c-a589-11e8-07c0-fd5c91b9da4a] to /root/.julia/compiled/v1.4/Gen/OEZG1_t5nDi.ji.
+Stacktrace:
+ [1] error(::String) at ./error.jl:33
+ [2] compilecache(::Base.PkgId, ::String) at ./loading.jl:1272
+ [3] _require(::Base.PkgId) at ./loading.jl:1029
+ [4] require(::Base.PkgId) at ./loading.jl:927
+ [5] require(::Module, ::Symbol) at ./loading.jl:922
+ ```
+
+ This can even happen when running inside a VM on macOS. While this bug is confirmed and reproducable, there is currently no solution for it, cmp. [issue 311](https://github.com/probcomp/Gen.jl/issues/311).