Appearance

Getting Started ​

This guide walks you through your first photometric redshift fit with Lazy.jl.

1. Generate a Parameter File ​

Lazy.jl uses TOML configuration files. Generate a starting template:

bash
lazy params > my_params.toml

This creates a fully commented example file with all available options.

2. Configure Your Parameter File ​

Open my_params.toml in a text editor. The file has four sections:

[io] – Input and Output ​

Point to your FITS catalog and choose an output location:

toml
[io]
    input_catalog = '/path/to/your/catalog.fits'
    output_file = '/path/to/results.fits'
    missing_data_format = 'nan'
    output_pz = true
    output_templates = true

[fitting] – Scientific Parameters ​

Choose a template set and configure the redshift grid:

toml
[fitting]
    template_set = 'sfhz'
    template_error = 'template_error'
    template_error_scale = 0.2
    igm_model = 'inoue14'
    nphot_min = 2
    sys_err = 0.05
    z_min = 0.0
    z_max = 6.0
    z_step = 0.01

Use lazy list-templates to see all available template sets.

[runtime] – Processing Options ​

For most datasets, the defaults are fine. Enable chunked processing for very large catalogs:

toml
[runtime]
    chunked_processing = false
    target_memory_gb = 0.5

[translate] – Column Mapping ​

This is the most important section to get right. Map Lazy.jl filter names to the flux and error columns in your catalog:

toml
[translate]
    f606w = {flux = 'f_f606w', error = 'e_f606w'}
    f814w = {flux = 'f_f814w', error = 'e_f814w'}
    f090w = {flux = 'f_f090w', error = 'e_f090w'}
    f115w = {flux = 'f_f115w', error = 'e_f115w'}
    f150w = {flux = 'f_f150w', error = 'e_f150w'}
    f200w = {flux = 'f_f200w', error = 'e_f200w'}
    f277w = {flux = 'f_f277w', error = 'e_f277w'}
    f356w = {flux = 'f_f356w', error = 'e_f356w'}
    f410m = {flux = 'f_f410m', error = 'e_f410m'}
    f444w = {flux = 'f_f444w', error = 'e_f444w'}

The keys (e.g., f606w) are filter nicknames – use lazy list-filters to see all available filters and their nicknames. The values must match column names in your input FITS catalog.

3. Run the Fit ​

bash
lazy fit -p my_params.toml

To use multiple threads for faster processing:

bash
lazy fit -p my_params.toml -t 8      # Use 8 threads
lazy fit -p my_params.toml -t auto   # Use all available threads

Lazy.jl will display a progress bar with estimated time remaining and memory usage.

4. Inspect the Output ​

The output FITS file contains multiple extensions:

  • SUMMARY: Best-fit redshifts, chi-squared values, confidence intervals, and model photometry for every object

  • PZ (if output_pz = true): Full redshift probability distributions

  • TEMPL (if output_templates = true): Template SEDs with IGM attenuation, enabling SED reconstruction

You can read the results in Python:

python
from astropy.io import fits

results = fits.getdata('results.fits', ext=1)
print(results['z_best'])    # Best-fit redshifts
print(results['z_med'])     # Median redshifts from P(z)
print(results['z_l68'])     # Lower 68% confidence bound
print(results['z_u68'])     # Upper 68% confidence bound

See Output Formats for full details on all output columns and how to reconstruct best-fit SEDs.

Next Steps ​

  • Configuration: Complete reference for all parameter file options

  • Templates: Learn about the available template sets

  • Filters: Browse the full filter database

  • Advanced Usage: Chunked processing, caching, resume, and more