This guide includes a Continuous Wavelet Transform (CWT), significance tests from based on Torrence and Compo (1998)
and Cross Wavelet Analysis (CWA) based on Maraun and Kurths(2004).
First steps:
cd PATH_TO_SAVE_WAIPY_IN_YOUR_HOME
git clone https://github.com/mabelcalim/waipy.git
cd waipy/
A local install can be done using the provided setup.py file:
python3 setup.py install
The installation path can be changed using the --prefix switch, e.g.:
python setup.py install --prefix $HOME/inst
Make sure to add the corresponding paths to your $PATH
and $PYTHONPATH
environment variables. Alternatively, if the --user switch can be used:
export PYTHONUSERBASE=$HOME/inst/pip_installs
export PYTHONPATH=$HOME/inst/pip_installs/lib/python2.7/\
site-packages/:$PYTHONPATH
export PATH=$HOME/inst/pip_installs/bin:$PATH
python setup.py install --user
If you plan on modifying the code, use the develop target in combination with the --user swich:
export PYTHONUSERBASE=$HOME/inst/pip_installs
export PYTHONPATH=$HOME/inst/pip_installs/lib/python2.7/\
site-packages/:$PYTHONPATH
export PATH=$HOME/inst/pip_installs/bin:$PATH
python setup.py develop --user
The first three lines should also be included in the $HOME/.bashrc file.
The following packages are required (tested on Debian Wheezy):
- python-matplotlib
- python-numpy
- libnetcdf-dev
- libhdf5-dev
- netcdf4 (https://github.com/Unidata/netcdf4-python.git)
- pip3 install numpy
- pip3 install matplotlib
- pip3 install netCDF4
- pip3 install pandas
- pip3 install scipy
As a tip, try to use jupyter !
- pip3 install jupyter
Check the prêt-à-porter.ipynb examples!!!
- https://github.com/mabelcalim/waipy/blob/master/Waipy%20Examples%20/waipy_pr%C3%AAt-%C3%A0-porter.ipynb
- https://github.com/mabelcalim/waipy/blob/master/Waipy%20Examples%20/cwa_pr%C3%AAt-%C3%A0-porter.ipynb
https://groups.google.com/forum/?hl=en#!forum/waipy-users-support