prov3it

my docker-compose: ``` version: "3.8" services: vpn: container_name: vpn image: dperson/openvpn-client cap_add: - net_admin environment: TZ: 'Europe/Amsterdam' FIREWALL: '' networks: - web read_only: true tmpfs: - /run - /tmp restart:...

> Hi, @xiaohulihutu, For single image inference, using only one GPU is what we expected. May I ask about the image size in this "cuda out of memory" case? >...

Seriously? Will this ever be fixed?

I'd like to see this feature as well. I've got thousands of scanned negatives which need subalbums to know in what physical folder i can find them.

Also failing on High Sierra (10.13.6).