32 lines
1.2 KiB
Markdown
32 lines
1.2 KiB
Markdown
|
# README #
|
||
|
|
|
||
|
|
This is HMCLET: a small extra HMC framework for ARES to allow sampling a bunch of model parameters together. It provides a self calibration step to estimate
|
||
|
|
the masses for the HMC.
|
||
|
|
|
||
|
|
### What is this repository for? ###
|
||
|
|
|
||
|
|
* Quick summary
|
||
|
|
* Version
|
||
|
|
|
||
|
|
### How do I get set up? ###
|
||
|
|
|
||
|
|
To include HMCLET please do a git clone of this repository in the `extra/` subfolder of ARES. For example:
|
||
|
|
|
||
|
|
cd $ARES/extra
|
||
|
|
git clone git@bitbucket.org:bayesian_lss_team/hmclet.git hmclet
|
||
|
|
|
||
|
|
Finally you can go to your build directory run `cmake .` to refresh the detected modules and run "make" again to finish the building
|
||
|
|
with HMClet included.
|
||
|
|
|
||
|
|
You can run `libLSS/tests/test_hmclet` to check that no error is triggered and verify the content of "test_sample.h5". It must contain a chain with 2 parameters
|
||
|
|
for which the first one oscillates around 1 with a variance of 10, and the other oscillates around 4 with a variance of 2.
|
||
|
|
|
||
|
|
### Contribution guidelines ###
|
||
|
|
|
||
|
|
You can submit pull requests to the BLSS team admin.
|
||
|
|
|
||
|
|
### Who do I talk to? ###
|
||
|
|
|
||
|
|
Guilhem Lavaux or Jens Jasche.
|
||
|
|
Check [our website](https://aquila-consortium.org/people.html) and the [wiki](https://aquila-consortium.org/wiki/).
|