Explore projects

This is the example how to deploy OKD python applications

Notebooks and related for the ML workshop at AIP

Lithium abundance fitting functions

try to build my own image

The test instance of daiquiri

Notebooks and related for the ML workshop at DESY

The docker related project for the pipeline

the pipeline repo for the ml_solar_plate project

A machine learning library to detect solar spots on APPLAUSE solar plates: https://www.plate-archive.org/applause/

orchest repo from ml_solar_plates

Excel Table

First demo for NFDI-PUNCH4 TA4

DEPRECATED. Please use the latest version from: https://gitlab-p4n.aip.de/punch-dev/punch-docs

the extract features steps: this step uses the threshold method to identify pixels that may be part of a sunspots

the label-features step: in this step the pixels that may be part of a sunspots are grouped in regions and each region is labeled as a feature, the number of feature per plate is computed on the fly.

The select-meaningful-features step: in this step we set the meaningfulness of each plates, and of each feature

The extract-features-properties step: in this step we compute the geometrical/morphological properties of each meaningful features

The modern pipeline code (prototype). Provide a config file, it builds the pipeline, setups a traceable environment, and run the pipeline on a K8s cluster (assuming access to some PV).