What are you working on? Any new ideas that you're thinking about?
Airplane AI - an offline first AI that is GDPR and NDA safe by architecture - totally 100% offline as the main value proposition, not a limitation
I wanted to get LLM feedback while writing without having the LLM suggest/write text for me, so I built https://www.writelucid.cc
We're building a repairable and fireproof ebike battery at https://infinite-battery.com :)
Economic simulation in Common Lisp.
supplementdex.com - we read (every study) to provide clinical decision support for practitioners wanting to learn about dietary supplements
tldr: we help you find good supplement
like you
Another thing I’m working on: homemade linen.
Right now I just germinated a 4x8 bed with flax for fiber. The plan is to grow it for 100 days or so and then harvest, dry, ret, dry, and spin. I need a lot more to do anything serious, but I think it’d be awesome to have a scarf that I made with linen I grew and harvested myself
[dead]
[dead]
[dead]
[flagged]
[dead]
[dead]
[flagged]
[dead]
[flagged]
[flagged]
[flagged]
[dead]
[dead]
[dead]
[flagged]
[dead]
[dead]
[flagged]
[dead]
[flagged]
[dead]
[dead]
[flagged]
[dead]
[dead]
[flagged]
[flagged]
[dead]
[flagged]
[dead]
[dead]
[dead]
[dead]
[flagged]
[dead]
[dead]
[dead]
[dead]
I just started on an open source and open weight supervised learning model to recognize japanese kanji characters drawn on the screen.
I have a working prototype written in Julia which is a very simple neural network. The input is in vector format so traditional convolutional neural networks don’t work out of the box but I swapped the convolution layer with a path simplification algorithm and it worked extremely well. Like 20 samples per character (from a set of only 5 hiragana during prototype phase) was enough to get 100% accuracy in a test collection of 5 samples per character after only 30 iterations of training.
I plan an working with free and open data, which I don‘t think exists for japanese kanji characters (at least not in vector format; KanjiVG only has one sample per character and I need dozens) so I also build a crowdsourcing web site to collect data from random people on the internet.
I am planning to run some more experiments with my prototype model before I release the crowdsourcing web page to an actual server though.
Model prototype: https://github.com/runarberg/kantoku-prototype
Crowdsource app: https://github.com/runarberg/kantoku-collector