Install error.

[donleo78]

Describe the bug
Im getting this error during fresh instalation process.

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
numba 0.57.1 requires numpy<1.25,>=1.21, but you have numpy 1.26.2 which is incompatible.
tensorflow-intel 2.12.0 requires numpy<1.24,>=1.22, but you have numpy 1.26.2 which is incompatible.

It happens only when fresh install it, on PC.
It works but slow, and after a little use it sends this message:

Exception in callback _ProactorBasePipeTransport._call_connection_lost(None)
handle: <Handle _ProactorBasePipeTransport._call_connection_lost(None)>
Traceback (most recent call last):
File "H:\roop-unleashed\roop-unleashed\installer\installer_files\env\lib\asyncio\events.py", line 80, in _run
self._context.run(self._callback, *self._args)
File "H:\roop-unleashed\roop-unleashed\installer\installer_files\env\lib\asyncio\proactor_events.py", line 165, in _call_connection_lost
self._sock.shutdown(socket.SHUT_RDWR)
ConnectionResetError: [WinError 10054] Se ha forzado la interrupción de una conexión existente por el host remoto

Anyway, amazing work. Thanks!

[C0untFloyd]

#525 (comment)

[donleo78]

Awesome, i see some changes when i run it today, it installed something, but i still get this error:

Downloading numpy-1.24.3-cp310-cp310-win_amd64.whl (14.8 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 14.8/14.8 MB 14.9 MB/s eta 0:00:00
DEPRECATION: torchsde 0.2.5 has a non-standard dependency specifier numpy>=1.19.*; python_version >= "3.7". pip 24.0 will enforce this behaviour change. A possible replacement is to upgrade to a newer version of torchsde or contact the author to suggest that they release a version with a conforming dependency specifiers. Discussion can be found at pypa/pip#12063
Installing collected packages: numpy
Attempting uninstall: numpy
Found existing installation: numpy 1.26.2
Uninstalling numpy-1.26.2:
Successfully uninstalled numpy-1.26.2
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
tensorflow-intel 2.12.0 requires numpy<1.24,>=1.22, but you have numpy 1.24.3 which is incompatible.
Successfully installed numpy-1.24.3
Launching App

Do i have to open a new thread? i just want to help you fix the errors, but i dont want to bother you.
Thanks!

[C0untFloyd]

No, I think this can be treated more or less like a warning. I don't think tensorflow-intel is used by me anywhere. And the app should be working, shouldn't it?

[lysxelapsed]

Numpy issues lead to pytorch issues in my experience. Not exactly sure to what extend pytorch is used in roop-unleashed nowadays though.
@C0untFloyd I'm guessing the installer overwrites a changed requirements.txt when running the installer a second time? Is there a way around that? Trying to help unexperienced users, which are using the installer, is a b****. This (and countless other cases) would be so easy to solve otherwise: adjust numpy in requirements.txt to 1.24, run the installer again.

[C0untFloyd]

I'm guessing the installer overwrites a changed requirements.txt when running the installer a second time?

No, the installer is hardcoded to that release (commit hash). The requirements won't be changed.

adjust numpy in requirements.txt to 1.24, run the installer again.

That's what I already did in the latest commit: 9907f1f

[lysxelapsed]

I'm guessing the installer overwrites a changed requirements.txt when running the installer a second time?

No, the installer is hardcoded to that release (commit hash). The requirements won't be changed.

Well then:
@donleo78 Open the requirements.txt file in your roop-unleashed folder, adjust the numpy version to 1.24, save the file and run the installer again.

adjust numpy in requirements.txt to 1.24, run the installer again.

That's what I already did in the latest commit: 9907f1f

Commit is 1.24.3, it's asking for 1.22 - 1.24 (and not a byte higher, python is so insufferable 🤣) - As to why it's 1.24 in his case and 1.24.3 in mine: Maybe different python versions?

[C0untFloyd]

Don't want to argue and you're most helpful but in his case this wasn't a breaking error anyway:

Successfully installed numpy-1.24.3
Launching App

I still have it running with numpy 1.26 something
I need to find out which of the 3rd party things imports tensorflow-intel (it's not used anywhere by me).

[lysxelapsed]

OMG reading is hard on my part🤦‍♂️ I always read Successfully UNinstalled numpy-1.24.3. Result of just glancing through it. Sorry for wasting everyone's time 🙈