David Black's Projects
Word-frequency for Ancestry job description and my résumé. https://stackoverflow.com/a/38178031/6505499
For now, a quick python script to set up a nice Excel document to check how far along I am with my classes.
Use for getting MS images from combined image files.
The repo to put @deeplizard YouTube tutorials/courses. Thanks to auto-suggestion for the repo name, GitHub!
Some nice data visualizations regarding Discrete Fourier Transforms. These are done in the Wolfram Language (the one used to power Mathematica). The PDF and PNG versions of the notebooks are the best to use for seeing the visualizations.
Make my environment the way I like it.
Work done in the AWS MLS Certification course at Udemy from Chandra Lingam. He's an awesome teacher. I don't own anything. Oh, and now, since I can't get the Jupyter Notebooks on AWS working, I'm just bring the notebooks here, and I'll do the re-typing here, too.
Python modules, scripts, classes, etc. that I use very often and want to have available quickly.
Things I use often, which are nice to have available without rewriting. Hooray code re-use.
The classic Machine Learning problem of recognizing digits from the MNIST dataset. Both single-hidden-layer approach and CNN approch.
Some good reference stuff.
My visualization of the Higgs Boson discovery data as well as other analysis and comments.
A repository to go along with a Physics StackOverflow question concerning some issues I'm having with analyzing some 4-lepton data from the 2011 part of the Higgs-hunt data.
Dumping the useful info, files, etc. So they don't weight down things, now that I'm doing a second try.
Word-frequency from job descriptions and my résumé, cover letter, etc. Note that an easy-to-navigate version is at the polished-5part-presentation branch.
Code & research description to be presented at the 2024 Family History Technology Conference. Research and Development on a tool to find document images where manuscript waste (a.k.a. binder's waste) was used to bind other documents. Info from reused manuscripts can relate to genealogy and/or MS studies, and is extensible to finding, e.g. repairs.
A little NLP fun: finding which words can be made using only the symbols of Mendeleev's Table (perhaps more commonly, though less justly) also known as the Periodic Table. The trick here is that we will use multiple languages. Even scripts (to the degree that they can be found), starting with Cyrillic.
Following the LinkedIn course by Zhongyu Pan called "Natural Language Processing with PyTorch". URL: https://www.linkedin.com/learning/natural-language-processing-with-pytorch
A great intro dataset for data exploration & visualization (alternative to iris).
A place to put links and pictures of projects I want potential employers to see
Tracking indicators for my health
Geting LoRA working with RWKV (v5 Eagle, probably, rather than v6 Finch). Large Language Model stuff. Get GPT performance from RNN-like LLM.
A request to contribute to the RWKV5-LM-LoRA project - specifically by helping with the issue, " Can you add the requirements.txt file? #4". I plan to look through the code for the modules used and build up a `requirements.txt` file until I can run everything.
Going through ZTM (Zero To Mastery) Google TensorFlow Developer Certification Course
Learning R
Learning R (arrghh).
Complete Guide to TensorFlow with Deep Learning with Python. Udemy
Things that I want available at a quick reference. Things that anyone can see (not related to any jobs). Quick instructions, lists of programs to install, etc.
A place to put some of my favorite visualizations and Machine Learning work, whether completely-guided from tutorials or completely-original, or anywhere in between.
For my personal website at bballdave025.com . For now, it only allows a small bit of HTML, and that's not even in full-HTML format.