Giters

kartben / artificial-nose

Instructions, source code, and misc. resources needed for building a Tiny ML-powered artificial nose.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Fails to build

flufftronix opened this issue · comments

commented

Hi, I couldn't get a build to compile with my own data from Edge Impulse, so I tried the /firmware directory straight from this repo, and am getting errors like this:

Log

Any ideas? Thanks!

commented

Interesting - thanks for reporting, and for including the logs :)
Looks like there was a recent change in the Seeed_Arduino_Linechart library that caused the regression. I have pushed a fix (which consists in sticking to the old version of the lib), could you please confirm it works for you?
Cheers!

commented

================================================= [SUCCESS] Took 67.79 seconds =================================================

There we go :) thanks again!

commented

Very good to hear! I would love to know more about the kind of project you're working on, btw :)

commented

Thanks! After some initial testing that didn't go as smoothly as I was hoping, I've had to put this on the backburner for awhile.. but the goal is to mine some potential for using gas sensors as a method of harm reduction drug checking.

I live in Philadelphia in the US. and there's an ever-increasing problem here with adulterants being added into the illicit drug supply; notably a veterinary tranquilizer called xylazine that's in most of the substances sold as dope here. Xylazine has never been tested or approved for use in humans and is leading to some really ghastly wounds and sores. Not to mention lowering someone's blood oxygen considerably, leading to loss of consciousness and a likely increase of vulnerability to opoid overdose. There aren't any reagent tests for it available yet, but I've been told it has a smell some people have learned to pick up on (the chemical contains a sulfur atom; I'm no chemist but I have some suspicions there!).

And while xylazine-adulterated dope has become the norm in Philly, it's just starting to show up in other places. So at some point, I'm hoping to at least be able to knock out some proof of concept testing. And in general I'm hoping that affordable sensors plus deep learning can help fill the gap between affordable but single use drug checking via reagents and test strips, and various types of spectroscopy and chromatography machines which are generally a lot more thorough and detailed, but cost more than most people make in a year.