5 ms·
Looking at this project and some of the alternatives provided, they seem to be quite memory hungry. I assume this is related to the ML model. Is there any way
by JD557 6y ago
Looking at this project and some of the alternatives provided, they seem to be quite memory hungry. I assume this is related to the ML model.
Is there any way to run this (or one fo the alternatives) in a "Lite" mode without any ML?
I see that Photoprism has a ton of flags, but it's not obvious how those affect memory usage.
- hanklazard 6y agoYeah, I think for users of Google Photos, the ML was a big part of it. Lots of other options without, I’ve used and like Lychee (1). (1) https://lychee.electerious.com/ https://lychee.electerious.com/
- olliebennett 6y agoPhotoPrism uses TensorFlow for the image classification / ML stuff (to add labels to your images), but it can be completely disabled via a flag (or toggle under advanced settings) before you begin indexing if you don't need it. PhotoPrism's ML object detection is no doubt not as powerful as Google's, but it seems to work pretty well based on the demo, and I know the devs are working on improving it frequently.
- cozzyd 6y agoyeah, 8 GB of RAM for a photo gallery seems completely insane...
- mceachen 6y agoPhotoStructure will run comfortably with 1gb RAM, and scales to very large (>1mm file) libraries. The ML models I'm loading for the next version may take a bit more RAM than that, but (as most features with PhotoStructure), you'll be able to selectively disable them to taste. https://photostructure.com/faq/why-photostructure/ https://photostructure.com/faq/why-photostructure/
- derneuere 6y agoWe are not sure why it needs that much RAM. We know that it has something to do, with the ML models. I am pretty sure we can optimize it in the future.