-
Notifications
You must be signed in to change notification settings - Fork 418
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Numpy 2.0.0 conflicting breaking Plugin HASS #2013
Comments
Same |
1 similar comment
Same |
Where exactly did you specify Pandas to cause this? Was it in an app? |
I used the GUI within HomeAssistant and a requirements.txt, both leading to the same error related to np.float64 / numpy 2.0.0. incompatibility. The workaround, so far, completely resolves this issue for me. I specified numpy==1.26.4 in the HA GUI and installed and imported other packages (i.e. scipy and PuLP) without any issues so far. |
Thanks for clarifying, and supplying the workaround. I'll leave this open for a while for other folks to see, I don't see that I can do anything to fix it. Numpy isn't an explicit requirement for AD, it seems to get pulled in by deepdiff, but if I add it to our requirements and pin it I might cause other issues later on when they fix it. Let me know if you see something different I can do to help. |
I use numpy extensively in appdeamon apps and get the same issue. Tried specifying an older version (numpy==1.26.4) in requirements.txt in the same directory as the appdeamon.yaml file, but this does not help. |
I finally figured out where to amend the config in the GUI. See picture in above post. |
Thanks! |
Thank you very much, I was looking into it for the last few days and couldn't figure out what made my sensors in HA stop working. |
HASS plugin is using the DeepDiff module , which will install numpy as a dependency appdaemon/plugins/hass/hassplugin.py -> Line 14 -> from deepdiff import DeepDiff The DeepDiff module is using float_ which was removed in numpy 2.0 release, which caused it to break and hence , also break the HASS plugin File "/usr/lib/python3.11/site-packages/deepdiff/helper.py", line 63 |
mmm, I can see that it was fixed in the Deepdiff module 10 months ago, so probably our fix should by updating the Deepdiff version rather than downgrading the numpy version |
appDaemon is using deepdiff==6.3.0 So we probably need to update to at least 6.6.0 to fix this isuse |
An alternative fix seems to be to specify deepdiff>7.0.0 in the appdaemon config. Appdaemon complains but loads OK |
Thanks for the research - we will update the deepdiff version and see how that works out.
…________________________________
From: fboundy ***@***.***>
Sent: Thursday, August 15, 2024 10:36 AM
To: AppDaemon/appdaemon ***@***.***>
Cc: Andrew Cockburn ***@***.***>; Mention ***@***.***>
Subject: Re: [AppDaemon/appdaemon] Numpy 2.0.0 conflicting breaking Plugin HASS (Issue #2013)
An alternative fix seems to be to specify deepdiff>7.0.0 in the appdaemon config. Appdaemon complains but loads OK
—
Reply to this email directly, view it on GitHub<#2013 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACLE7JH5OP7NOSA3SHOBXDDZRS4GPAVCNFSM6AAAAABJVSOESKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEOJRGQYDONZXGY>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
OK, I bumped deepdiff to 7.0.1 in dev - can anyone let me know if this fixes the issue? |
Just wanted to mention that adding matplotlib also can lead to this error. Adding |
Thanks - without actually checking I believe we just updated dev to 8.0.0 so we should be past this when the next version of AD releases in a few weeks
|
What happened?
When loading the most recent Pandas package, Numpy 2.0.0 also gets loaded. This breaks the import of HASS. See log file. Workaround of specifying numpy==1.26.4 resolves the issue.
Version
4.4.2
Installation type
Home Assistant add-on
Relevant log output
Relevant code in the app or config file that caused the issue
No response
Anything else?
No response
The text was updated successfully, but these errors were encountered: