aganse Goto Github PK
Name: Andy Ganse
Type: User
Location: Seattle
Blog: research.ganse.org
Name: Andy Ganse
Type: User
Location: Seattle
Blog: research.ganse.org
messy/quick wrapper implementations of some ML models, just for personal use
Inverse problem solving tools, in progress
Miscellaneous data science / machine learning Jupyter notebooks
A simple gradio-based web-app that connects to a database, displays images, and allows to enter json tags.
A ready-to-run Docker container setup to quickly provide MLflow as a service, with PostgreSQL, AWS S3, and NGINX
My own tweaks on the awesome and popular 🧛🏻♂️ dark theme for tmux
Flow-based invertible neural networks implemented with Keras, Tensorflow, and Tensorflow Probability.
DBSCAN image-clustering based automated reading and decryption of fairy cipher glyphs from 2nd Artemis Fowl book
C programs and shell script to read/filter data and calculate soundspeed profiles from NOAA's World Ocean Database 1998 (the version I used for many of my analyses; newer data is available at OCL website, with some filters, but these programs offer more in-depth filtering.)
gpt_client: a Python-based OpenAI GPT client that can be run as a terminal/readline based CLI or as a browser-based web-app
A comparison/demonstration of five predictive filters and smoothers (KF, KS, LKF, EKF, EKF2) based on an expansion of two examples in Gelb's "Applied Optimal Estimation", comparing and contrasting them in both linear and nonlinear cases in the measurements as well as in the dynamics. See my linked webpage discussing the problem.
Reinforcement learning applied to Bagh Chal board game, via OpenAI's Gym toolkit
fit a 3d box to a point cloud
Calculate Tikhonov-regularized, Gauss-Newton nonlinear iterated inversion to solve the damped nonlinear least squares problem (Matlab code).
MiniSim was a Java2-based test demonstration for a networked Navy simulation client. The simulation networking component itself can't be shared with the public, but the client is separable from the simulator so this fun GUI part can be shared. For quick kicks, see screenshot.png and download/run MiniSimDemo.jar (via "java -jar MiniSimDemo.jar").
MultiRegressLines runs multiple-line (up to three) segment regression on a dataset of X,Y data. This Java version of my code has some remaining math issues - till those are fixed please use the Matlab version for actual analysis (MultiRegressLines.matlab). For both, my linked webpage explains this topic of "multi-phase linear regression".
The Matlab scripts in MultiRegressLines.matlab run a two-line-segment regression on a dataset of X,Y data, and analyze the uncertainties in that regression via Monte Carlo (since it's a nonlinear problem).
A minimal version of a JSci-like package from back before that was a viable option. So this is just a legacy library of mine that RayDemo, SpheRayDemo, and MultiRegressLines depend on.
My updated versions of some old Ptolemy Plot (PTplot) class files, used in my dataTools, RayDemo, SpheRayDemo, and MultiRegressLines packages.
A ready-to-run Python/Tensorflow2/MLflow/Docker setup to train models on GPU and log performance and resulting model in MLflow.
Docker-compose setup for quick-prototyping a Postgresql database via Docker, serving SchemaSpy result via Nginx
Train and test a yolov3 model that detects/tracks rabbits in our yard via grainy nighttime IR photos
RayDemo is an interactive, graphical, geophysical raytrace applet, particularly useful for demonstration/tutorial purposes. Its focus is geophysical such as sound/earthquake waves propagting through the ocean or land, as opposed to light reflections between objects (a different popular use of raytracing for graphics needs).
SpheRayDemo is a "spherical raytrace demonstration" of modeling raytraced ion acoustic waves propagating in the Sun. See my webpage on helioseismology for more. Bonus: this app's plots look like Spirograph art!
Fortran + Matlab/Octave to read/process NASA spacecraft trajectories from publicly available NAIF SPICE files for proposed Europa Clipper mission (thanks to NASA NAIF for assistance). While this same trajectory data is available independently online, this code provides a framework for processing trajectory data in spacecraft and lunar coordinates.
Accelerating new GitHub Actions workflows
Matlab tutorial demonstration scripts on beamforming concepts, used to create the results shown on my beamforming tutorial webpage.
A Vim plugin to browse your MLflow parameters and metrics from within Vim instead of (or in addition to) the MLflow webapp GUI.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.