This program uses python 3. Dependencies are listed in requirements.txt. It is recommended you install in a virtual environment. I recommend using the built-in venv module for python3, as described in the docs.
- Prepare Python alias
I strongly recommend using git bash if you are in a Windows environment. Run git bash as administrator. You will need to set an alias for python in Windows.
You can now proceed to the directions for a Unix-like Environment
$ alias python3='winpty python.exe'
- Create virtual environment
$ python3 -m venv .venv # create a virtual environment with a clean python version. The virtual environment is created in the directory `venv/` - Start Virtual Environment
$ source .venv/bin/activate # enter/start the virtual environment
- Install Dependencies in Virtual Environment
(.venv) $ pip3 install -r requirements.txt # install dependencies for the specific script into the virtual environment's python.
You are now ready to run the program.
To run the algorithm, use the following format:
$ python3 GP_Main.py path/to/input.csv path/to/output.txtIf you would like to print all output possible for the program, include the argument --verbose with the above in any position after GP_Main.py.
Your input file should be a csv containing columns for your covariates and one column titled presence that contains 1's and 0's where 1 represents a success ("species presence") and 0 represents a failure ("species psuedo absence"). All other columns will be included as covariates. It is important to balance your dataset so that the number of 1's and 0's in presence are approximately equivalent.
Your output files will be written as a text file formatted using indentations so you can visualize the tree. It will also include the fitness of the tree.
To force quit the process in a Unix-like environment, use Ctrl-C.