pydeb is a Python package for building and analyzing Dynamic Energy Budget models. It supports phylogenetic inference of DEB models for new species, simulation of growth, reproduction and survival, and Bayesian inference with the Adaptive Metropolis algorithm. It is designed for high performance and can analyze 100,000 parameter sets in under 2 minutes on most workstations.
You need a Python distribution with NumPy and Cython.
To run the examples, you additionally need jupyter and plotly.
If you are starting from scratch, we recommend you install a Python distribution such as Anaconda, which comes with most of these packages preinstalled. You can then install any remaining ones with conda, e.g., conda install plotly.
If you have the prerequites described above, install pydeb with:
python -m pip install <PYDEB_DIR> --user
Notes:
pythonis the name of the your Python interpreter. Replace it if necessary. For instance, some systems have Python 3 installed aspython3. If you want to install pydeb for this distribution, usepython3 -m pip install <PYDEB_DIR> --user.- If you run the above from the pydeb top-level directory (the one that contains this README file),
<PYDEB_DIR>would be. - The above installs in your user-specific site packages directory. To install system-wide, omit
--user. But this may require administrator/sudo permissions.
pydeb comes with a number of Jupyter Notebook examples.
To use these, run jupyter notebook from the examples subdirectory.