Encourage people to pick the master branch

[myankele]

The latest tag is really out of date and doesn't work.
Each error can be researched at lengths, but they were already fixed in master.
Or we can tag the current tip ?

[baudneo]

Zm ml is right around the corner, docker based and hardware accelerated. Need testers, zmes will still function though if people prefer that.

[myankele]

That's great news, I can contribute to testing.
I am only suggesting that in the meantime, since internet is jam packed with ZMES information, and ZMES is still being actively fixed up, we should change that official document to tell people to grab ZMES master tip instead of a 2 year old tag.

[baudneo]

There are far more problems than that. DLib 19m24 had API changes and won't work for face recognition. Opencv 4.7.0 has an issue that won't allow for ML inference. Maybe it's fixed in 4.7.1, if you don't compile opencv from source and download it from pip then you need to use the rolling release from februaury.

There's a couple other things as well that I can't recall and we've only been doing the bare minimum to get zmes to install.

@connortechnology might merge your PR, I have no say or control over that.

[jantman]

Zm ml is right around the corner, docker based and hardware accelerated. Need testers, zmes will still function though if people prefer that.

I just stumbled here because I moved and am doing a completely fresh ZM install, along with trying to get everything running in Docker. I've spent about 2 days trying to get ES working in Docker. @baudneo do you have a pointer to this "ZM ML"? I'd be overjoyed to test if it's already in Docker.

[baudneo]

Zm ml is right around the corner, docker based and hardware accelerated. Need testers, zmes will still function though if people prefer that.

I just stumbled here because I moved and am doing a completely fresh ZM install, along with trying to get everything running in Docker. I've spent about 2 days trying to get ES working in Docker. @baudneo do you have a pointer to this "ZM ML"? I'd be overjoyed to test if it's already in Docker.

It is delayed a bit ATM. I will start reworking on it and hopefully January 2024 I will have a release for beta testing for the ZM community. I am running it at home and it is ridiculously fast and rock solid. I just need to implement some logic changes and refactor the code base.