- create object and classes instead of using bare functions.
- Track all the used libraries and update in requirements.txt and environment.yml.
- Update MANIFEST.in and setup.py to include auxiliary files.
- create and upload the whole python package to PyPi(test.pypi).
- Find a suitable freezing package for 1-click excecutable creation. Probably can use PyInstaller as suggested here
- Test the v0.1 executable for windows.
- The package is up on PyPi, so use pip to install
pip install geomexp
The installation of dependencis shouldn not be necessery. Use only if you run into trouble with dependencies.
Once the package is installed then just import it using import geomexp.ml
and you can then run the Muller-Lyer experiment as follows.
geomexp.ml.ml_exp()
The experiment will start by asking some basic details about the subject. The esc
key will pause the experiment, and q
will quit the expeiment. The gathered data will be stored in a folder called "data", at the location where you are running the terminal.
The environment.yml
file can be used by conda to create a environment which can run the package. The first line of the file will be the name of the environment. In this case it is psychopyP2N
.
If you have conda installed(or download here) then use the following command to create an environment on top of which the package can run.
conda env create -f environment.yml python=2.7
The command for activating the environment just created is
conda activate psychopyP2N
For further information read here.
The method with pip
and requirements.txt
is not currently working.
conda install --name psychopyP2N --yes --file requirements.txt
pip install -r requirements.txt
conda env export > environment_droplet.yml
conda env create -f environment.yml
install requires
- gcc
- build-essentials
- binutils
- python2- setuptools
- python2-subprocess32
- conda install psutil
- conda install wxPython
- conda config --add channels anaconda to be updated