Playing with stable diffusion on your own machine is great.
Stable diffusion, prompt and parameters:
machines slaving humans in a dystopic future, dramatic lighting, cinematic lighting, moody, greg ratkowski, turner, goya, trending on artstation, highly detailed -s100 -W512 -H512 -C7.5 -Ak_lms -U 4.0 -F -S2237964052
An essay on Communist era jokes.
These questions have real potential. I have added them as good food for thought. They are at the company level, though some could apply at department or team level.
This is a very good overview of what
means being a staff engineer.
It is in jest, of course.
Interesting development. Going for Snowflake?
Is it me, or is this putting all the burden on data leaders?
How HashiCorp uses RFC (
Requests for Comments) documents. How they are structured, length, consistency.
Good rules. They make sense. They are not cool though.
I’ve been following the writings from
John on and off for more than 10 years. Interesting to see how he sees his work.
Just like we dedicate our time and energy toward a career, committing ourselves to a “serious leisure” activity is one of the keys to achieving a fulfilling life, he says.
It mentions the 5 learning gaps, which I think I have seen in some learning/expertise book but I had forgotten about. And they are a neat concept.
Brilliant. Tempted to go for D!
This is an interesting view on how a company is structured (and where DevRel sits).
Between comments in the thread and links available, you can certainly get it running. It is not superfast (5 seconds/iteration more or less) but gets the work done. If you want upscaling you’ll need to edit
ldm/gfpgan/gfpgan_tools.py and remove some conditions there.