====== Polpy & threeML======

Below are the instructions to install __Polpy__ and __threeML__. Everything is to be done inside the Python environment.

===== Prerequisites =====

**<color #ed1c24>The process of installing Prerequisites varies with different operating systems.</color>** 

**1. If you are using a MacOSX-arm64 system, please follow:**

1.1 Install miniconda3

First, we need to install miniconda Do this through:

<cli>
mkdir -p ~/miniconda3
curl https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-arm64.sh -o ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm ~/miniconda3/miniconda.sh
</cli>

1.2 Conda Environment

Set up a Python virtual environment using conda (install Anaconda if you don't have it already). Do this through 

<cli>
conda create -n polpyenv python=3.9
</cli>

Activate your conda environment 

<cli>
conda activate polpyenv
</cli>

When you finish your analysis, you can exit the environment by

<cli>
conda deactivate polpyenv
</cli>

You can deactivate it through "conda deactivate". Install pip inside conda:

<cli>
conda install conda-forge::pip
</cli>

1.3 Installing Necessary Packages

Install the necessary packages through the following commands:

<cli>
conda install -c conda-forge numpy pandas matplotlib scipy
pip install --upgrade pip setuptools wheel
conda install -c conda-forge hdf5
pip install cmake
conda install -c conda-forge gfortran
conda install conda-forge::pygmo
conda install conda-forge::numba
conda install -c conda-forge pytables
</cli>


**2. If you are using a Linux or MacOSX-Intel system, please follow:**

2.1 Use __pyenv__ for a Virtual Environment

Creating a virtual environment allows you to install packages safely without interfering with the system Python. 

You can use __pyenv__ for this purpose (or __virtualenvwrapper__, instructions in this [[https://virtualenvwrapper.readthedocs.io/en/latest/]]).

You can install __pyenv__ by executing this line in your terminal:
<cli>
curl https://pyenv.run | bash
</cli>

After installing the package, append the auto-loading lines printed on the terminal to ‘~/.bashrc’ or ‘~/.zshrc’, and source the files.  

2.2 Establishing a Virtual Environment

We suggest fixing Python 3.9 for this work (to avoid deprecation of dependencies).

With __pyenv__, you can install Python 3.9 by:

<cli>
pyenv install 3.9 
pyenv virtualenv 3.9 polpyenv 
pyenv activate polpyenv
</cli>

Once you are done with all your analysis, you can deactivate the virtual environment with (don't do it now but when you finish your analysis):

<cli>
pyenv deactivate
</cli>

2.3 Installing Necessary Packages

Install the necessary packages through the following commands:

<cli>
pip install --upgrade pip setuptools wheel
pip install numpy pandas matplotlib scipy
pip install pygmo
pip install numba
</cli>


**<color #ed1c24>Once the prerequisites are satisfied, you can thereafter follow the steps below commonly for different systems.</color>**

===== Clone Repositories =====

Create a dedicated directory for the whole project/environment. Navigate to the folder containing the whole project:

<cli>
mkdir workshop
cd workshop/
</cli>

Thereafter, cone the GitHub repos through:

<cli>
git clone https://github.com/sujaymate/astromodels
git clone --branch polpy_dev --single-branch https://github.com/sujaymate/threeML/
git clone https://github.com/JohannesBuchner/MultiNest/
git clone https://github.com/JohannesBuchner/PyMultiNest
git clone --branch patch-1 --single-branch https://github.com/hellohancheng/polpy/
</cli>



===== Installing Astromodels =====

Assuming you are in the "workshop" directory. Install the __astromodels__ through:

<cli>
cd astromodels/
pip install .
cd ../
</cli>

===== Installing threeML =====

Assuming you are in the "workshop" directory. Install the __threeML__ through:

<cli>
cd threeML/
pip install .
cd ../
</cli>

===== Installing MultiNest =====

Assuming you are in the "workshop" directory. Install the __MultiNest__ through:

<cli>
cd ../MultiNest/build/
cmake ../
make
sudo make install
cd ../..
</cli>

===== Installing pyMultiNest =====

Assuming you are in the "workshop" directory. Install the __ pyMultiNest__ through:

<cli>
cd PyMultiNest
pip install .
cd ../
</cli>

===== Installing polpy =====

Assuming you are in the "workshop" directory. Install the __ polpy__ through:

<cli>
cd polpy
pip install .
cd ../
</cli>

===== Running example script to test Polpy & threeML =====

Assuming you are in the "workshop/polpy" directory. You can run the example script through:

<cli>
cd examples
python polar_GRB170114A.py
</cli>

===== Optional =====

Thereafter, if your script running complained about the incompatibilities among Numpy, Astropy and other packages, you can downgrade your Numpy by doing this:

<cli>
pip install numpy==1.26.4 --force-reinstall --no-cache-dir
</cli>

Add the following lines to your .bash_profile file (or similar).

<cli>
vim ~/.bash_profile
</cli>

Inside add the following lines.

<cli>
export OMP_NUM_THREADS=1
export MKL_NUM_THREADS=1
export NUMEXPR_NUM_THREADS=1
</cli>

Save and exit the file. Then do:

<cli>
source ~/.bash_profile
</cli>