Queryable
Queryable copied to clipboard
mazzzystar
Synology(群晖) NAS really needs this feature to search photos.
Many photos are stored in the Synology NAS, but when there are too many, it becomes hard to find specific ones. I hope there can be a solution for this. Thank you
@shoreson In order to make it readable for everyone, I have translated it into English.
I haven't used Synology before, and I'm wondering what would be the best way to provide it? For instance, through a plugin or a Docker image?
Alternatively, in the original app, I can read photos from the Synology through the address, and search both on my local machine and in the Synology when searching.
Alternatively, in the original app, I can read photos from the Synology through the address, and search both on my local machine and in the group view when searching.
got it. I'm not familar with Synology plugin dev, just checked out the document, seems to be similar with making a docker image, would like to have someone to try : )
Alternatively, in the original app, I can read photos from the Synology through the address, and search both on my local machine and in the group view when searching.
got it. I'm not familar with Synology plugin dev, just checked out the document, seems to be similar with making a docker image, would like to have someone to try : )
Thanks for your efforts and investigation - thanks to the user who request it ( started this issue ) If it's possible It would be nice to try it too. I'm actually using a fork of MeowAI-Paddle for fully auto tagging photos over the Synology API. But yopur solution sound better for me. I've also tried to port a packege from DSM6 to DSM7 - in the end it was a success so maybe we can start to work together.
@akaydev-coder @nodis @shoreson Is it OKay to just provide the Python Docker with build_index.py and search.py so someone can develop based on it?
And you can find the implementation from here: https://github.com/immich-app/immich/discussions/1613
@mazzzystar thank you for your cooperation and support. I had a completely wrong approach to how Queryable works. I assumed that it analyzes the photos like MeowAi/PaddleAi and tagged them according to relevance. MeowAi/PaddleAi indirectly tag the photos and write directly to the running photos database via Python. The process also runs once. I would argue it is easier to build the Docker image than to create a SynoPackage. A corresponding Swift Docker image for Synology already exists. Only limited executable on ARM.
I have to admit that I have been very overzealous. This is going to be difficult from what I know. iO Photos is the next problem: it would be ideal to include the Photos Synology folder in iOS Photos and Queryable accesses it instead of storing all photos on the device itself.
Constructing the appropriate user interface ... phew xD
I have to work hard on all this, I don't have that much vacation ^^
Your app is nova - as advertised extremely fast and effective. MeowAi/PaddleAi has cost me nerves that it starts at all - MeowAi with 80 tags is defacto useless. PaddleAi was even more difficult to get running (Ubuntu VM + Py3.8) both variants slow and the recognition rate is average.
@akaydev-coder Offering a python docker image maybe the best way to run on Synology I guess.
+1 for this issue.
Is it OKay to just provide the Python Docker with
build_index.pyandsearch.pyso someone can develop based on it?
I have a Synology NAS but I'm not familar with Synology plugin and swift/app. I'm also a rookie of Docker and ML & DL, but I'd like to try to make some glue code to make this works.
BTW, Thank you for developing this awesome app. ❤️
@chAwater thx, and actually, you could consider adding a paid feature to your application, which might help it sustain itself longer.
@chAwater thx, and actually, you could consider adding a paid feature to your application, which might help it sustain itself longer.
haha, sounds like a sad story..
need Synology support also. For quick development and deployment, I think a docker image is beeter.
