lpy_tools

This package contains algorithms that can be used with L-Py to automate particular tasks. Most of the tools are meant to be used with the L-Py python API in a python file.

Getting started

Install conda package

You can install the openalea.lpy & lpy_tools conda packages in an environment named lpy with python 3.9 as follows:

conda create -n lpy python=3.9
conda activate lpy
conda install openalea.lpy -c fredboudon -c conda-forge
conda install lpy_tools -c mosaic -c fredboudon -c conda-forge

⚠️ You many need to install OpenGL system dependency. On Ubuntu sudo apt install freeglut3-dev.

Install from sources

To create a new conda environment (will be named lpy_tools) and install the required dependencies:

conda create -f conda/environment.yml

To install dependencies in an existing environment, first activate it (with conda activate <myenv>), then:

conda env update -f conda/environment.yml

Build conda package

You will need the conda-build package to build packages. Install it in the base environment with:

conda install conda-build

Then you can install the library in develop mode with:

python -m pip install -e .

⚠️ This should be done in the activated conda environment!

To build the lpy_tools conda packages, in the base environment from the root folder, run:

conda build -c conda-forge -c fredboudon conda/.
# Convert linux package for OSX-64:
conda convert -p osx-64 ~/miniconda3/conda-bld/linux-64/lpy_tools*  --output-dir ~/miniconda3/conda-bld/
# Convert linux package for WIN-64:
conda convert -p win-64 ~/miniconda3/conda-bld/linux-64/lpy_tools*  --output-dir ~/miniconda3/conda-bld/

⚠️ This should be done from the base environment!

You will need the conda-verify & anaconda-client package to upload packages. Install them in the base environment with:

conda conda-verify anaconda-client

After a successful build, to upload the packages you can run:

anaconda login  # do this once
anaconda upload --user mosaic --label main ~/miniconda3/conda-bld/linux-64/lpy_tools*.tar.bz2
anaconda upload --user mosaic --label main ~/miniconda3/conda-bld/osx-64/lpy_tools*.tar.bz2
anaconda upload --user mosaic --label main ~/miniconda3/conda-bld/win-64/lpy_tools*.tar.bz2

Official documentation on how to upload packages here.

Conda useful commands

Purge built packages:

conda build purge

Clean cache & unused packages:

conda clean --all

Module snapshots

Explanations

It is frequently useful to produce simulation outputs where some variables are changed, but using the exact same camera settings, color-maps etc.

For this, the snapshots module contains functions that make it possible to launch L-Py from a python file and create images corresponding to different runs of L-Py where one or two parameters of the L-Py models are changed from the outside of the model. These parameters must be declared as extern(...) global variables in the L-Py model file:

extern(diam = 0.3)
extern(angle = 30)

The function browses a list of parameter pairs defined by the user and produces sub-figures that are all assembled into a single figure, with controlled camera settings.

Quick example

We provide a simple example for this module in examples/fractal_snapshots.py. To use it, move to the example folder and run the example with:

cd examples
ipython fractal_snapshots.py

Module point_sampler

Explanations

This module provides functions that make it possible to compute a point cloud corresponding to the point sampling of a L-Py geometric model output. Labels can be attached to each point by attaching a label to each individual module label using a dictionary.

This function can be called within the python shell, or as a python function.

Quick example

We provide a simple example for this module in examples/??.py. To use it, move to the example folder and run the example with:

cd examples
ipython ??.py