#/
32 messages · Page 1 of 1 (latest)
The default is 512MB RAM with only 1 core
I don't think tensorflow would be able to run in that.
I got no idea actually, I have 4 core 4096MB as my maximum
Also this is just my instinct, it looks like your app is just not getting any resources
Have you tested the application on your own machine to check how much it needs
Can you add logs to your functions to see where your code reaches before it times out
Can you add the requirements to a barebone function, import them and see if it runs or not. Don't execute anything just imports
Let me test out some versions of tensorflow myself to see which ones build properly
Yep
Can you show me how you got the python 3.11 version. I never deployed one on that runtime
Let me try this out
Do they follow the appwrite functions structure and have a response?
I am not sure what the problem is
The only error I got was opencv
I fixed it by using opencv-python-headless instead of opencv-python
It even ran on a 0.5 CPU 512 MB environment
I did not get any errors after that
Kind of off topic but any word on sites availability? I have been waiting since the announcement day. I don't want to use another provider to just serve universal links to my app in which I exclusively used appwrite for everything 😅
Yes
6 mins
Yep 10 secs approx
I didn't execute anything just the starter function with the imports that u sent
I just added this
Just had appwrite from the beginning
Alright
It did install version 2.19.something of tensorflow from what I saw from the build logs check the build logs to see what it installed for you
Your requirements have a >= for tensorflow so I thought any version is fine?
Is it with a bare starter kit?
Where exactly are you executing it from
Try using the console
Also in the main file add the imports
It takes a bit
Have you increased the timeout?
How did you fix it
I am heading to the office right now, I will try to come up with something for your timeout problem during lunch
Seems reasonable
/