Skip to content

Spectrometer data processing from field devices

License

Notifications You must be signed in to change notification settings

UofU-Cryosphere/spectro-dp

Repository files navigation

example workflow

spectro-dp

Spectrometer data processing library from field devices.

Current supported devices

  • ASD field spectrometer

Command Line Interfaces

After the environment is set up and the library installed, it provides the below listed command line interfaces.

A detailed help for each command is provided when giving the --help option. All programs will prompt for required parameters if not given with the call.

asd_albedo

Calculate the snow albedo from a sequence of ASD measurements.

Sample call

asd_albedo -in /path/to/measurements/ -fp file_prefix -up 0 -down 10

asd_reflectance

Calculate the reflectance from surface and white reference ASD measurements.

Sample call

asd_reflectatnce -in /path/to/measurements/ -fp file_prefix -rs 0 -wrs 10

asd_white_reference

Utility to inspect a sequence of white reference measurements.

Sample call

asd_white_reference -in /path/to/measurements/ -fp file_prefix -wrs 10

Installation

This library was developed with a conda environment, using the supplied environment.yml and environment_dev.yml.

Conda

Follow the instructions of the official documentation to get started with conda.

Create a conda environment

Command line interface execution

To use the command line interfaces, the execution environment needs to be set up using the environment.yml:

conda create --file environment.yml

Library development

For development and running the tests, both .yml (environment.yml and environment_dev.yml) files need to be used for setup:

conda create --file environment.yml --file environment_dev.yml

Library installation

This step needs to be done for both, execution and development environment.

  1. Clone the repository locally
  2. Change path to inside the library
     cd spectrod-dp
  3. Install the Library
    pip install -e .

Development

All commands are covered by a test suite under the test directory. Make sure that all tests pass once you completed code changes.

Running tests

Running the test suite from inside the project root directory

pytest tests/

About

Spectrometer data processing from field devices

Resources

License

Stars

Watchers

Forks

Packages

No packages published