README

quicklensHEAL

Basically the same as D.Hanson's quicklens but uses HEALPix alm2map_spin instead of ECP 
=====

This package contains code for generating simulated CMB maps with 
lensing effects, as well as estimating the lensing potential from 
an observed sky. It has routines for:
     * simulating lensed and unlensed CMB maps in both the 
       full (curved sky) case, as well as the flat-sky limit.
     * inverse-variance filtering a flat sky map, to downweight 
       noisy modes for power spectrum or n-point function analysis. 
     * running the full suite of temperature+polarization quadratic 
       estimators on a filtered sky map (as well as a framework for 
       easily implementing other quadratic anisotropy estimators).

The focus for estimators is on quick, flat-sky routines, however 
code for evaluating full-sky estimators and their noise spectra 
is included as well.

Scripts in the 'examples' directory give quick demonstrations of 
major use cases. Primary examples include:

     * examples/plot_lens_reconstruction_noise_levels.py
        Plot lens reconstruction noise power spectra for the 
	TT, EE, TE, TB and EB estimators, assuming a toy 
	experiment with Gaussian beam and white instrumental noise. 
	Plots are made for both full-sky and flat-sky calculations 
	for comparison.
       
     * examples/cinv/test_cinv_teb.py
	Run an inverse-variance filtering operation on a flat-sky 
	map using preconditioned congjugate descent.

     * examples/lens/make_lensed_map_flat_sky.py
        Generate a set of lensed CMB maps in the flat-sky limit.

     * examples/lens/make_lensing_estimators.py
        Runs a set of quadratic CMB lensing estimators on lensed 
	maps, and plots their power spectra as well as cross-
	correlation with the input lensing potential.

This code has the following dependencies:
     * numpy (required)
     * scipy (optional, used for smoothing in 
       	     one of the plotting routines)
     * pypar (optional, used for coordinating parallel 
       	     generation of simulations and lensing analysis). 

The code can be run either from this directory, or installed by 
running "python setup.py install".

The code is primarily written in Python, although some low level 
spherical harmonic transform (SHT) routines are written in Fortran 
for speed. In order to use these SHT routines, the code must be 
comiled. This is done automatically when installing, however if 
running the code without installing it needs to be built with

"python setup.py build_ext --inplace"

Depending on system, you may need to specify a fortran compiler. 
For example

"python setup.py build_ext --inplace --fcompiler=gnu95"