3KIDNAS · JS/DCP pipeline
Galaxy Fit
Source code adapted from https://github.com/NateDeg/3KIDNAS
Fits one HI data cube's kinematics via a Nelder–Mead tilted-ring model, entirely in JS/DCP — no Fortran, no Python. Runs as two separate DCP jobs: an initial/anchor fit, then a bootstrap-uncertainty loop sized by how many realizations you ask for. Nothing here reads or writes anything outside your own machine except the two DCP dispatches themselves.
file://. Serve this folder locally instead, e.g. from a terminal in this same
directory: python3 -m http.server 8000, then open
http://localhost:8000/index.html.
DCP Account
Paste your DCP API key to view and manage the job in the DCP Portal, and secify the compute groups that determine which workers can pick up your jobs.
Leave joinKey blank for the public group. Not specifying one deploys to Global (public).
1Initial fit
The anchor fit every bootstrap realization is measured against.
No data on hand?
Object name auto-fills from the cube filename. PA/Inc estimates are manual entry.
Advanced — fixed pipeline configuration
Defaults transcribed from this pipeline's own config templates. Same values for every galaxy unless changed here.
| Ring property | Constant across rings | Fixed at initial value |
|---|
2Bootstrap uncertainty loop
Configurable any time -- "Run bootstraps" unlocks once the initial fit above has converged. Cost scales with the realization count.
Join the compute cluster
Scan to turn your phone or laptop into a worker for the compute group this job is running on -- more workers join mid-job, and it visibly speeds up.