QualityAssurance

Quality Assurance software for NISAR

For the upcoming NISAR mission, eight types of L1/L2 data products will be generated. This Quality Assurance (QA) software is designed to look at the data products produced one at a time. For each product, the QA code can:

• Verify the metadata matches the product spec
• Generate metrics, a PDF report, and a summary CSV describing the quality of the product
• Run CalTools processes on RSLC products
• Produce a browse image PNG with sidecar KML file for that product

Minimum PreRequisites:

See environment.yaml for required packages.

Operating Instructions:

Installation

Step 1) Install the conda package manager.

Step 2) Clone the nisarqa repo and cd into the top level directory.

git clone https://github.com/isce-framework/nisarqa.git
cd nisarqa

Step 3) Create a conda environment with the correct packages for QA.

By creating a new environment using the environment.yaml file, the required packages and their versions will all be resolved by conda from the conda-forge channel, and thus remain consistent with each other. (Mixing package managers has led to import errors during testing.) The environment.yaml defaults to naming the new conda environment nisarqa.

conda env create -f environment.yaml
conda activate nisarqa  # activate the new environment

Step 4) Install

For standard installation, run:

pip install --no-deps .

For develop mode, run:

pip install --no-deps -e .

To test installation, try:

nisarqa --version
nisarqa -h
nisarqa dumpconfig -h

Caution

--no-deps is necessary for installation due to the isce3 dependency. Otherwise, isce3 would need to be removed from pyproject.toml, causing that dependency list to fall out of sync with environment.yaml's list.

Running the QA Code

Run the QA code from the command line using a runconfig.yaml file.

Because the QA code is uniquely written for each product type, each product type has a unique subcommand.

See nisarqa -h and e.g. nisarqa rslc_qa -h for usage.

Example commands to process QA:

nisarqa rslc_qa <rslc_runconfig.yaml>
nisarqa gslc_qa <gslc_runconfig.yaml>

By default, QA outputs the majority of log messages to the log file only. To additionally stream the log messages to the console, use the verbose flag. Example:

nisarqa gslc_qa <gslc_runconfig.yaml> -v
nisarqa rslc_qa <rslc_runconfig.yaml> --verbose

Step 5) Set up pre-commit (Optional for users) (Required for contributors)

QA repo uses a pre-commit workflow to ensure consistent code style for developers. Configuration options are found in the pyproject.toml and .pre-commit-config.yaml files.

Install pre-commit and check if it was correctly installed:

conda install -c conda-forge pre-commit
pre-commit help

Install and set up the git hook scripts:

pre-commit install  # stdout: pre-commit installed at .git/hooks/pre-commit

Now pre-commit will run automatically on git commit.

Some useful commands:

pre-commit run --all-files  # run on all files (not just files in commit)
git commit --no-verify -m "message"  # skip pre-commit entirely
SKIP=flake8,black git commit -m "foo"  # temporarily disable flake8 and black

Runconfig Template w/ default parameters

Because the QA code is uniquely written for each product type, each product also has a unique runconfig yaml file template and default settings.

The runconfig is where a user specifies the filepath to the input NISAR product, which workflows to run, what units to use for a given metric, and so forth.

To get a product's example runconfig file that has been populated with the default parameters, use the dumpconfig subcommand.

See nisarqa dumpconfig -h for usage.

Example commands:

nisarqa dumpconfig rslc
nisarqa dumpconfig rslc --indent 2                # change the indent spacing
nisarqa dumpconfig rslc > my_rslc_runconfig.yaml  # save runconfig to a file
nisarqa dumpconfig gcov

Expected Outputs

For each NISAR product, if all workflows are requested via the workflows parameter in the runconfig, the QA code will generate six output files and store them in the directory specified by qa_output_dir in the runconfig:

1. BROWSE.png - RGB browse image for the input NISAR product
2. BROWSE.kml - geolocation information for BROWSE.png
3. REPORT.pdf - graphical summary PDF containing histograms, low-res images of the input datasets, etc.
4. STATS.h5 - statistical summary HDF5 file containing computed quality metrics, the datasets used to generate the plots in REPORT.pdf, the QA processing parameters, etc.
5. SUMMARY.csv - a high-level PASS/FAIL check summary
6. LOG.txt - textual listing of errors encountered during QA processing

These file names are hardcoded; they are not configurable via the runconfig.

Copyright

Copyright 2024, by the California Institute of Technology. ALL RIGHTS RESERVED. United States Government Sponsorship acknowledged.

Export Classification

This software may be subject to U.S. export control laws. By accepting this software, the user agrees to comply with all applicable U.S. export laws and regulations. User has the responsibility to obtain export licenses, or other export authority as may be required before exporting such information to foreign countries or providing access to foreign persons.

License

This software is licensed under your choice of BSD-3-Clause or Apache-2.0 licenses. The exact terms of each license can be found in the accompanying LICENSE-BSD-3-Clause.txt and LICENSE-Apache-2.0.txt files, respectively.

SPDX-License-Identifier: BSD-3-Clause OR Apache-2.0