This would fit most best as a “Show HN:” post :)
I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
11–20 of 129 posts
Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#12Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#13Does it work for porn collections too?
Asking the important questions
Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#14 M1 Max CPU is an ARM/SoC, comparable to an 11th gen Intel i9
Do I have it right? Would Windows ARM performance be similar for those cpu?ref: https://www.cpubenchmark.net/compare/4585vs4245/Apple-M1-Max...
Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#15Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#16Does it work for porn collections too?
Last time I tried whisper, it hallucinated an elaborate conversation from sounds of slapping and moaning and it took minutes to spit every single line of it.
Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#17Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#18Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#19I was surprised to learn that the M1 Max CPU is an ARM/SoC, comparable to an 11th gen Intel i9 Do I have it right? Would Windows ARM performance be similar for those cpu? ref: https://www.cpubenchmark.net/compare/4585vs4245/Apple-M1-Max...
- "unified" ram makes all the system ram available as VRAM - dedicated ai coaccelerator thingy
Both of these reasons allow the apple silicon chips to crush conventional cpus in these kind of AI model workload stuffs
No idea about what the windows arm stuff is capable of. I know they use Qualcomm snapdragon chips though.
Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models
#20Does it work for porn collections too?
You'll need a lora for this, porn content rejection is heavy. Or you'll need a abliterated model, not sure if vision also works. You might want to add something like yolo finetune to detect scenes + face recognition too.