guilhem_lavaux/gl-website-deployer
0
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update website

Browse source
This commit is contained in:
Guilhem Lavaux 2024-11-19 08:02:04 +01:00
parent 4413528994
commit 1d90fbf296
6865 changed files with 1091082 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

2
generic/abyss.rst Normal file
View file

@ -0,0 +1,2 @@
ABYSS
=====

6
generic/calendar.rst Normal file
View file

@ -0,0 +1,6 @@
Aquila calendar
===============
.. raw:: html
<iframe width="100%" height="600px" src="https://cloud.aquila-consortium.org/apps/calendar/embed/icSZdSYQNpo4jZmi"></iframe>

61
generic/code.rst Normal file
View file

@ -0,0 +1,61 @@
Public software
===============
Here is a list of all public software that I wrote and that are available on this website:
.. raw::
<div class="features">
<section>
<span class="icon solid major fa-code"></span>
<h3><a href="/v.php/flints">FLINTS</a></h3>
<p>Check out this <a href="/v.php/flints">page</a></p>
</section>
<section>
<span class="icon solid major fa-code"></span>
<h3><a href="/v.php/cosmotool">CosmoTool</a></h3>
<p>This is a patch work of homegrown cosmology tool. Some details are available <a href="/v.php/cosmotool">here</a> and on the bitbucket <a href="https://bitbucket.org/glavaux/cosmotool/">repository</a></p>.
</section>
<section>
<span class="icon solid major fa-coden"></span>
<h3><a href="/v.php/galaxexplorer">GalaxExplorer</a></h3>
<p>This is an ancient software but you can still check it out at the following <a href="/v.php/galaxexplorer">page</a></p>.
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3><a href="/v.php/borg">BORG 1.x & 2.x</a></h3>
<p>ARES/BORG <a href="https://bitbucket.org/bayesian_lss_team/ares/">repository</a>
</p>
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3><a href="/v.php/borg">BORG 3.x</a></h3>
<p>The <a href="https://bitbucket.org/aquila-consortium/borg/">main repository is here</a>. The public version of ARES/BORG (a restricted portion only including the linear modeling) is available <a href="https://bitbucket.org/bayesian_lss_team/ares_public">here under a CECIL/GPL license</a>. Some other aspects are described my <a href="/v.php/borg">BORG page</a>.
</p>
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3>Web visualisation of MCMC</h3>
<p>borg_webvisu allow to run interactive visualisation of MCMC chain in the web browser, find it in the <a href="https://bitbucket.org/glavaux/borg_webvisu/src/master/">repository</a>.
</p>
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3><a href="/v.php/splotch">PySplotch</a></h3>
<p>A version of splotch bound to python and with some tweaks. <a href="https://bitbucket.org/glavaux/pysplotch/src/master/">Here</a>.
</p>
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3><a href="/v.php/miniauction">MiniAuction</a></h3>
<p>A minified implementation of the MAK algorithm for large scale structures. Check the page for more details.
</p>
</section>
<section>
<span class="icon solid major fa-scroll"></span>
<h3><a href="/v.php/abyss">ABYSS/BStatSampler/a></h3>
<p>An early toolbox (prefiguration of BORG) for statistical analysis of CMB signals with MCMC methods.
</p>
</section>
</div>

16
generic/cosmotool.rst Normal file
View file

@ -0,0 +1,16 @@
CosmoTool
=========
Install
-------
For installing the pre-build python packages it is possible to just do
.. code-block::
pip install cosmotool
At the time of writing (Dec 22nd 2022), the binary package were prebuilt for linux for python 3.6 to python 3.9.
To build from source you have to obtain the software from bitbucket using git.

33
generic/flints.rst Normal file
View file

@ -0,0 +1,33 @@
FLINTS
======
Overall presentation
--------------------
FLINTS is an interpolation algorithm of fields sampled on a sphere. We assumed originally that the field corresponds to a band limited Gaussian random field but it is also optimal in the least-square sense for any field on the sphere.
We propose here a C++/OpenMP implementation of this algorithm. It uses the specificities of HEALPix pixelization for speed optimization. The code is mostly self-documenting through comments.
Software requirements
---------------------
* CMake (http://www.cmake.org) for building FLINTS
* HEALPix (http://healpix.sourceforge.net/)
* GSL (http://www.gnu.org/software/gsl/)
* Optionally GenGetopt (http://www.gnu.org/software/gengetopt/)
Download
--------
* Version 1.0 is here
* Version 1.1-beta3 is here
* Bitbucket repository: https://bitbucket.org/glavaux/flints
Documentation
-------------
The documentation is included in the tarball of the 1.1 release. It consists into a user manual and the corresponding refereed article.
License
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License. This GPL license is available at the address http://www.gnu.org/licenses/.

31
generic/galaxexplorer.rst Normal file
View file

@ -0,0 +1,31 @@
GalaxExplorer
============
Introduction
------------
GalaxExplorer is a real-time 3D visualization tool aimed originally at visualizing spectroscopic redshift galaxy catalog. It is more a hobby project than an attempt to do a full featured visualization tool (like Paraview).
Features
* handle large number of particles (tested up to ~64 millions particles on a 4 GB memory laptop);
* support for a number of point representation (points, spheres, tree hierarchical, textured), surface representation, and unstructured mesh representation with triangles (using Delaunay tesselation);
* Lagrangian transport of surfaces, with the surface representation mode.
* multiple skyboxes for orientation dependent background;
* real transparency through front peeling algorithm and GLSL shaders;
* support for tracing the trajectory of particles (currently require a large memory to load all particles and all timesteps, but online loading is planned);
* automatic movie capturing;
* remotely controllable through a D-Bus interface (a Python binding is included in the source package, data can be directly uploaded from python);
* snapshoting of the current state and reloading.
Requirements
------------
* QT4 toolkit
* HealPix, for skybox support
* OpenGL >= v2.1, some functionalities may require a higher version.
Getting GalaxExplorer
---------------------
GalaxExplorer is not anymore in development at the moment. It is available publicly on this bitbucket repository: https://bitbucket.org/glavaux/galaxexplorer

45
generic/icgen.rst Normal file
View file

@ -0,0 +1,45 @@
ICGen
====
Description
-----------
ICgen is a generator of initial condition for N-body simulations. It is based on MPI and FFTW 2. It is able to handle a theoretically unlimited number of constraints for generating constrained Gaussian random field.
Package content
---------------
After unpacking the source package, you have a number of files and subdirectories unpacked in the directory ICgen-X/ (X being the current version code). There are current two subdirectories: "src" and "samples". In "src", you will find the source code of ICgen. In "samples", there are a few sample configuration files that you can look at to get familiar with the parameters.
Building instructions
ICgen uses an autoconf/automake architecture for building. On standard linux distribution it is usually sufficient to run:
.. code-block:: bash
./configure
make
These two commands checks the availability of a working compiler, MPI, GSL and FFTW2. In general you do not have to do anything. However, if those packages are installed in non-default locations you may have to specify extra arguments to configure, in that case please have a look at the INSTALL file.
If everything compiles fine, you have an executable "potin_mpi" in the "src" subdirectory.
Running ICgen
-------------
For running and generating initial conditions, please execute:
.. code-block:: bash
mpirun -np X ./potin_mpi CONFIGURATION_FILE
with X the number of nodes on which you want to execute the generator. The format of the configuration file is detailed here.
Documentation
-------------
The documentation for the configuration file is on this page.
Download
--------
The current public version of ICgen is 0.1-15122009. The current private version is 0.2. The public version can be downloaded from `here <http://www2.iap.fr/users/lavaux/software/tgz/icgen-0.1-15122009.tgz>`_.

10
generic/miniauction.rst Normal file
View file

@ -0,0 +1,10 @@
MiniAuction is a slimmed version of the full MAK package. It is written in C++ and has support both for serial and MPI parallelization. The package comes with documentation and samples for the two algorithms.
This package does not come with the mesh element generator. These belongs to the full MAK package that will be released later on.
The current version is 1.0. The code can be downloaded here. There are no previous public versions. Please acknowledge the use of this code using the following statement:
This work makes use of the MiniAuction package version 1.0 written by Guilhem Lavaux.
This software has been developed with the financial support of the ANR research gr ant OTARIE (ANR-07-BLAN-0235).

11
generic/projects.rst Normal file
View file

@ -0,0 +1,11 @@
Projects
========
On this page I collect relevant research projects and code for my daily activities:
* BORG (https://bitbucket.org/aquila-consortium/borg/)
* ArrayFire (https://arrayfire.org/)
* JaX (https://jax.readthedocs.io/en/latest/)
* NBodyKit (https://github.com/bccp/nbodykit)
* Paraview (https://www.paraview.org/)
* Topological Toolkit (https://topology-tool-kit.github.io/)

4
generic/science.rst Normal file
View file

@ -0,0 +1,4 @@
Science
=======
Writing in progress...

78
generic/splotch.rst Normal file
View file

@ -0,0 +1,78 @@
PySplotch
=========
Rationale
---------
Splotch is a software for visualizing simulation originally written by Klaus Dolag and Martin Reinecke. The original website is
https://wwwmpa.mpa-garching.mpg.de/~kdolag/Splotch/.
PySplotch is a trimmed down version, alongside a python binding. I removed the CUDA acceleration and allowed for more generic data exchange between the rendering core and the user.
Notably there is now a Python binding and a number of helper object to control the camera, the transition between datasets and the generation of a large number of images for video making.
The source code may be found here: https://bitbucket.org/glavaux/pysplotch/.
Example
-------
.. code-block:: language-python
import _splotch
import numpy as np
Np=20000
w,h=800,800
x = _splotch.particleArray(Np)
x[:] = np.random.rand(Np,3)
#x[0] = [0.5,.5,.5]
x.setIntensity(0.01)
x.setRadius(0.005)
x.setValues(np.random.rand(Np))
cmap = _splotch.SplotchColormap(num_types=1)
cmap.addValue(0, 0, (0, 0, 0))
cmap.addValue(0, .5, (.5, 0.5, 0))
cmap.addValue(0, .5, (0, 0.5, 0.5))
cmap.addValue(0, 1, (1, 0., 0))
theta=np.pi/4
r = 2
for i,phi in enumerate(np.arange(0,2*np.pi, 0.05)):
campos = _splotch.SplotchCamera([r*np.cos(phi)*np.cos(theta)+0.5,r*np.sin(phi)*np.cos(theta)+0.5,r*np.sin(theta)+0.5], [0.5,0.5,0.5], [0,0,1])
import _splotch
import numpy as np
Np=20000
w,h=800,800
x = _splotch.particleArray(Np)
x[:] = np.random.rand(Np,3)
#x[0] = [0.5,.5,.5]
x.setIntensity(0.01)
x.setRadius(0.005)
x.setValues(np.random.rand(Np))
cmap = _splotch.SplotchColormap(num_types=1)
cmap.addValue(0, 0, (0, 0, 0))
cmap.addValue(0, .5, (.5, 0.5, 0))
cmap.addValue(0, .5, (0, 0.5, 0.5))
cmap.addValue(0, 1, (1, 0., 0))
theta=np.pi/4
r = 2
for i,phi in enumerate(np.arange(0,2*np.pi, 0.05)):
campos = _splotch.SplotchCamera([r*np.cos(phi)*np.cos(theta)+0.5,r*np.sin(phi)*np.cos(theta)+0.5,r*np.sin(theta)+0.5], [0.5,0.5,0.5], [0,0,1])
```