Mert
Mert
Superseded by #16613
Anyone feeling adventurous (and who has 35gb of free space on their disk) can try this image: `ghcr.io/immich-app/immich-machine-learning:pr-16613-rocm`. You'll also need to add these lines to the machine learning service...
Are you sure this isn't a case of the container reaching resource limits / running out of memory? Also @redaws00, it's unlikely that an issue with your GPU has the...
You can post the machine learning logs with `docker logs immich_machine_learning`. In general, Docker on Windows can be convenient, but it's also much less stable than Docker on Linux. Issues...
That's pretty weird. There should be no difference between what the job does on manual queue vs on upload, and no difference in what it sends to the machine learning...
The likely reason is that the model has to be compiled to OpenVINO format the first time it's loaded, which (unfortunately) blocks the ML server from responding to any requests...
Typically a few minutes, at most maybe 15 in a very constrained environment with a large model.
A Pixel 3A is a quite old phone. I wonder if your specific issue is that a slower swiping animation "throttles" how quickly you can swipe between assets, masking the...
At least on iOS, I know that #13426 is when this issue started for me and reverting it fixed the issue.
The linked issue doesn't seem related to this PR. Also, the edited image is lossy and loses some qualities of the image, like HDR. I think we should make that...