Installation

Follow the base package installation below. Using the package manager miniconda is highly recommened.

Base package

Create a conda or virtual enviroment and activate it (optional, but recommended).

conda env create --name photometry python=3.11
conda activate photometry

python -m venv ./photometry_env
source ./photometry_env/Scripts/activate # for MacOS / Linux
photometry_env\Scripts\activate # for Windows

Install from PyPi or GitHub. Installing from GitHub will install the latest version.
```
pip install PhoPro
```
```
pip install git+https://github.com/ggolde/PhoPro.git
```

You can update the package with by including the --force-reinstall flag in the pip install command.

FMM support

Follow the next step after the basic installation to add support for the advanced analysis method Functional Linear Mixed Modeling. It is highly recommended that you install PhoPro in a conda enviroment if you want FMM support.

The setup can be a pain but the FMM analysis framework can be very insightful. For more information about FMM for photometry data see the original paper by Dr. Loewinger.

Install all R packages avaliable from conda:

conda activate photometry
conda install -c conda-forge r-base=4.4.3 r-essentials
conda install -c conda-forge r-lme4=1.1_35.5 r-mass=7.3_64 r-mvtnorm=1.2_6 r-gridextra=2.3 r-rfast=2.1.0 r-matrix=1.7_2 r-arrangements r-here r-devtools r-rcurl

Then run the install_FLMM.R script with your conda enviroment active.
```
Rscript envs/install_FLMM.R
```

If you get an error like Error: C17 standard requested but CC17 is not defined for any of your packages try creating the file ~/.R/Makevars and add the following to it depending on your platform:
```
CC17 = clang                # for MacOS
CC17 = gcc                  # for Linux and Windows
C17FLAGS = -O2 -std=gnu17   # for all
```

Install the fastFMM Python wrapper with:
```
pip install fast-fmm-rpy2
```
With your conda enviroment active verify rpy2 is working properly with:
```
python -m rpy2.situation
```

If you get a and error similar to Error importing in API mode: ImportError("dlopen(..._rinterface_cffi_api.abi3.so, 0x0002): Library not loaded: .../lib/libRblas.dylib, you will have to set the enviroment variable R_HOME within your conda enviroment then unintall and reinstall rpy2.
```
conda activate photometry
conda env config vars set R_HOME="$(R RHOME)"
conda deactivate
conda activate photometry
```
```
pip uninstall rpy2 -y
pip uninstall rpy2-rinterface -y
pip uninstall rpy2-robjects -y
conda install -c conda-forge rpy2
```