JWST Calibration Pipeline Testing

This is the documentation for calibration-pipeline-testing-tool.

Installation

Create a JWST pipeline environment to install into and install pytest-html

conda create -n test_jwst --file http://ssb.stsci.edu/releases/jwstdp/0.7.8/latest-osx
source activate test_jwst
pip install pytest-html

git clone https://github.com/STScI-MESA/calibration-pipeline-testing-tool.git
cd calibration-pipeline-testing-tool
python setup.py install

Basic Usage

To setup tests, you specify test input files in a JSON file, with an entry for each step. You do not need to provide input for every step. The below example shows all currently available steps. Any of the step names can be omitted and tests associated with that step will be skipped. Multiple FITS files can be supplied for any given step and the tests will be repeated for each supplied file.

{
    "dq_init": [
        "dq_init_input.fits"
    ],

    "saturation": [
        "saturation_input.fits"
    ],

    "superbias": [
        "superbias_input.fits"
    ],

    "dark_current": [
        "dark_current_input.fits"
    ],

    "refpix": [
        "refpix_input.fits"
    ],

    "linearity": [
        "linearity_input.fits"
    ],

    "rscd": [
        "rscd_input.fits"
    ],

    "lastframe": [
        "lastframe_input.fits"
    ],

    "jump": [
        "jump_input.fits"
    ],

    "ramp_fit": [
        "ramp_fit_input.fits"
    ]
}

Then from the command line simply run

test_pipeline --config confg.json

This will produce a summary.html file with the test results as well as plots if any are produced. These files are saved based on the input filename and will be overwritten on subsequent runs, so it is advisable run the test suite in it’s own directory. Note that this file and the associated plots will be saved in the current directory so it may be useful to run test_pipeline in a new directory.

Contributing

If you would like to contribute tests to the package see the Developer Guide.