Graphical user interface¶
If you are starting with raw CRNS files and little or no programming experience, begin with the beginner's guide. It covers installation, configuration detection, the first GUI run, calibration checkpoints, and export in one sequence.
Neptoon provides a Streamlit-based graphical user interface for users who want to process CRNS data without writing Python scripts.
Follow the installation guide for environment choices. The GUI uses the same YAML as the CLI: the configuration overview explains the file split, and the config wizard can create or update starting files.
Install the GUI extra¶
The GUI is currently maintained in neptoon tools GUI
The graphical user interface is currently maintained as a separate application,
neptoon tools GUI. Use it online
at https://neptoon.web.app.ufz.de/ without installing
anything, or install it on your own computer as described in the README of the
neptoon-tools-gui repository.
Installing this GUI directly through the gui extra of neptoon will follow here soon.
The GUI is an optional dependency set:
For an existing pipx installation, update with:
Start the GUI¶
The command starts a local Streamlit server and prints a browser URL, typically http://localhost:8501. Open that URL if your browser does not open automatically.
Typical workflow¶
- Select or prepare a sensor configuration YAML.
- Select a process configuration YAML.
- Choose the raw or preprocessed data location referenced by the config.
- Run processing.
- Download the processed data, figures, logs, and PDF report from the output folder.
The resulting folder layout is described in Saving data, with report options in PDF output. For repeatable command-line runs using the same YAML, see Using configuration files.
Notes¶
- The GUI uses the same configuration models as the CLI and Python API.
- If a GUI run fails validation, fix the YAML first and rerun; validation errors are usually more useful than downstream processing errors.
- For repeatable production processing, keep the YAML files under version control and use the CLI or Python API for automated reruns.
Roving page¶
The GUI also has a dedicated Roving page for mobile CRNS surveys. Upload a
roving YAML together with the local files it references. The page parses the
same RovingConfig used by the CLI/Python API, applies validated runtime
overrides, executes RovingProcessor, and previews captured interactive HTML
products. The controls are deliberately separated into 2D map, 3D terrain,
and Exports so that presentation settings do not get mixed with file-output
settings.
The 2D and 3D previews remain separate standalone HTML documents. The host mounts only the currently selected preview rather than putting both into eager tabs; this avoids keeping an invisible WebGL terrain scene alive while the 2D map is shown. The currently selected HTML can be downloaded directly. For GUI runs, the Streamlit page is the authoritative presentation-control surface by default: the browser-side 2D control panel, Leaflet layer selector, and 3D settings rows are suppressed so the same setting is not exposed twice. Enable Also show controls in generated HTML to keep any combination of the 2D map controls, 2D layer selector, and 3D terrain controls in the generated page as well. Changes made inside those generated HTML controls affect that page only; they are not written back into Streamlit or the saved configuration. Navigation and 3D export buttons remain available regardless of that choice.
The GUI therefore exposes the presentation choices before rendering, including a primary map provider plus the providers available to the 2D HTML layer selector, visible 2D layers, colour/interpolation/spatial-smoothing settings, 3D terrain appearance, terrain and rover palettes/ranges, track/point/Wurmi rover display, point size or track width, DEM smoothing, terrain-only vertical exaggeration, distance profile, interpolation-overlay opacity, footprint Q63/Q86 mode and opacity, water, buildings, and roads. Standalone HTML generated outside the GUI keeps its browser controls by default. Once a processed result is available, pure presentation edits are applied by re-rendering the map/terrain from that result; they do not repeat the scientific roving workflow. Processing-affecting changes still clear the cached result and require a new run.
For GIS use, the page also provides Download full-resolution rover KML for GIS from the processed survey. This download is independent of the browser's 3D trajectory thinning and contains the rover track plus point attributes, timestamps and elevation where available. Generated tables, maps and bundles remain available as normal output downloads.
Install the map/terrain extras needed by the outputs you enable, for example:
The GUI does not contain a separate roving calculation path: scientific and
spatial processing remains in RovingProcessor.