https://github.com/axsk/generators
Early work on the AJC and coherence (2019/2020) also with weather data
https://github.com/axsk/generators
ajc coherence julia pcca scrapbook sqra weather zib
Last synced: 26 days ago
JSON representation
Early work on the AJC and coherence (2019/2020) also with weather data
- Host: GitHub
- URL: https://github.com/axsk/generators
- Owner: axsk
- Created: 2021-09-14T10:44:43.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-10-29T10:15:45.000Z (over 4 years ago)
- Last Synced: 2024-03-24T06:10:17.054Z (about 2 years ago)
- Topics: ajc, coherence, julia, pcca, scrapbook, sqra, weather, zib
- Language: Jupyter Notebook
- Homepage:
- Size: 12.7 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Generators
Experiments around the EAMC and space-time committors / coherence.
The state of this repository is pretty messy and is left here for archival use.
The Augmented jump chain code has been rewritten for python and julia in `cmdtools` and `birthdeath`
Notable features: Coherence of meteorology data, PCCA+, Bickley Jet
## Contents
### Source
- Galerkin discretization of the Augmented Jump Chain (called Embedded augmented markov chain (EAMC) during that time): `src/galerkin.jl`
- EAMC reconstruction of the Koopman operator by space-time committor: `src/committor.jl`
- EAMC reconstruction of the Perron-Frobenius operator by jump activity: `src/perronfrobenius.jl`
- Gillespie like simulation of trajectories of the EAMC: `src/gillespie.jl`
- Opimization of space-time committors: `src/main.jl`
- PCCA+ implementation: `src/pccap.jl`
- Some interactive (?) Makie visualization: `src/plot.jl`
### (Nonautonomous) Models
- 4 Cell barrier on-off switching: `barrierswitch.jl`
- The bickley jet: `bickleyjet.jl`
- Fluid flow governed by vectorfield: `continuity.jl`
- Overdamped Langevin via SQRA, also shifting and rising doublewells: `langevin.jl`
- 1d sqra: `mysqra.jl`
### Notebooks
- `example*`: EAMC plots of different systems
- `bickley*`: bickley coherence, plots not saved
- `weather-inspect`: look at the weather data
- `weather-kmeans/pcca`: clustering of the eamc for coherent sets
- `weather-optimize`: clustering via optimization criterion