crowsonkb / v-diffusion-pytorch

v objective diffusion inference code for PyTorch.
MIT License
715 stars 108 forks source link

Request for argument #9

Open bmc84 opened 2 years ago

bmc84 commented 2 years ago

Hi, Firstly, thank you for releasing your amazing work!

This isn't an issue, but a request for an argument to be added to give the ability to save the image every x steps. Similar to the "display_every" option in the collab version, but if the value was set to 50 then instead we get something like "output0_step-250.png", "output0_step-300.png" etc.

Hope this isn't too much trouble. I know it's probably simple but it's above my Python skill level which is basically zero :)

Thanks again!