The Teach-It collection's progress has moved at a snail's pace, but it is still moving along. The challenge hasn't been creating a minimum viable product (MVP). Rather, engineering the project for speed, scale, modularity, automation, and aesthetic have been the major hurdles. If the reader remembers the early - and basic - web based version I made available some time ago, I'm sure it's clearer now why I've prioritized those attributes.
On the technical side of dissemination, I've shifted to the following tools:
The machine learning models powering Teach-It are still Python-based (TensorFlow / Keras / scikit-learn / etc.). I don't see that changing anytime soon, but I have been experimenting with MATLAB's model creation tools for another project. MATLAB (R2022b) hasn't completely convinced me, but there do appear to be some very bright spots in going down that route that I'll be keeping an eye on.
Anyway, things are still a work in progress, but the next update shouldn't take 9 months.
[ April 18, 2024 Update: Oh, the joys of being wrong... ]
Working on my Teach-It Collection has been quite a learning experience. In fact, I've learned so much that I'll essentially have to start all over!
My initial assumption was that my implementation would be so straightforward and simple that it could all fit in "one place". Well - as with many things in life - there is a level of simplicity to the project as well as a level of straightforwardness, but unassailable nuances here and there have made me realize that splitting everything up would make my world a bit better!
The key problems up to this point have essentially been the following:
With those points in mind, I plan on aesthetically simplifying - and technically complicating - the collection to get it where I would like it to be.
So, watch out for a smoother, snappier Teach-It collection experience that's easier on the eyes coming soon*!
*The sort of soon that may take a while...
This website uses cookies. By continuing to use this site, you accept our use of cookies.