Skip to main content

Mac (Apple silicon) Installation Guide

1. Check Command Line Tools​

Before proceeding, ensure that your Command Line Tools are up-to-date. Run the following in your terminal:

xcode-select --install

This will prompt you to install or update the necessary development tools.

2. Install Dependencies via Homebrew​

To install the required dependencies, use Homebrew. If Homebrew is not installed, you can install it by running:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

After installing Homebrew, use the following commands to install the necessary packages:

brew install python@3.11  # This installs Python 3.11 and pip3
brew install make gcc openblas

Set GCC as the Default cc Compiler​

If GCC is not set as your default cc compiler, create a symlink to ensure it's used:

# Create a symbolic link to use GCC as the default cc compiler
# Note: Adjust the GCC version if different (e.g., gcc-12)
cd $HOMEBREW_PREFIX/bin && ln -s gcc-12 cc

Verify: After this, check that cc points to the correct compiler:

which cc

Open a new terminal window if necessary for the changes to take effect.

3. Set Up Environment for Compilation​

Now, set the required environment variables for the compilation process:

export PKG_CONFIG_PATH="$HOMEBREW_PREFIX/opt/openblas/lib/pkgconfig:${PKG_CONFIG_PATH}"
export OWL_CFLAGS="-g -O3 -Ofast -funroll-loops -ffast-math -DSFMT_MEXP=19937 -fno-strict-aliasing -Wno-tautological-constant-out-of-range-compare"
export OWL_AEOS_CFLAGS="-g -O3 -Ofast -funroll-loops -ffast-math -DSFMT_MEXP=19937 -fno-strict-aliasing"
export EIGENCPP_OPTFLAGS="-Ofast -funroll-loops -ffast-math"
export EIGEN_FLAGS="-O3 -Ofast -funroll-loops -ffast-math"

If you want, you can now remove the cc symlink created earlier.

4. Install Dependencies via Opam​

You need to install Opam, the OCaml package manager. If Opam is not installed, follow the installation instructions here: Opam Installation Guide

Once opam is installed, create a switch for OCaml 5.0.0:

opam update
opam switch create 5.0.0
eval $(opam env)

Next, install the required OCaml packages:

opam install pyml toml lwt owl ocamlformat.0.24.1 dune linenoise

5. Install Miking and TreePPL​

The following script installs Miking and TreePPL compilers along with the required packages:

# Initialize environment
eval $(opam env)

# Clone and install Miking
git clone -n https://github.com/treeppl/miking.git
cd miking
git checkout 24505bd
make install

# Go back to the original directory
cd ..

# Clone and install Miking-DPPL
git clone -n https://github.com/treeppl/miking-dppl.git
cd miking-dppl
git checkout 680ea76
make install

# Go back to the original directory
cd ..

# Clone and install TreePPL
git clone https://github.com/treeppl/treeppl.git
cd treeppl
make install

# Go back to the original directory
cd ..

# Clone and install TreePPL-Python
git clone https://github.com/treeppl/treeppl-python.git
cd treeppl-python
pip install -e .
pip install matplotlib seaborn

This script will create four directories (miking, miking-dppl, treeppl and treeppl-python) where you execute the script and instruct you how to set up the environment variables. If you don’t plan to work on TreePPL development, you can safely remove the miking and miking-dppl directories after the installation:

rm -rf miking/ miking-dppl/

6. Customize Your Environment​

To ensure your environment is set up correctly every time you open a terminal, add the necessary environment variables to your ~/.zshrc file:

Edit your ~/.zshrc file:

nano ~/.zshrc

Then, add the following lines at the end of the file:

## Customizations for TreePPL
eval $(opam env)
export PATH="$HOME/.local/bin:$PATH"
export MCORE_LIBS="coreppl=$HOME/.local/src/coreppl/"
export MCORE_LIBS="$MCORE_LIBS:treeppl=$HOME/.local/src/treeppl/"

Save and close the file (in nano, press CTRL + X, then Y, and ENTER).

To apply the changes, either close and reopen the terminal or run:

source ~/.zshrc

7. Test Your Installation​

To verify that the installation worked correctly, restart your terminal session and run:

tpplc

You should see the manual page for the tpplc compiler.

To further verify the setup, you can run the example coin.py located in the treeppl-python directory:

cd treeppl-python/examples
python3 coin.py

If you are using a graphical terminal, a plot showing the inferred coin distribution should appear. If not, you can open the generated image file coin_outcomes_plot.png:

open coin_outcomes_plot.png

Once you have verified the output, you can delete the image:

rm coin_outcomes_plot.png