how to use GPU

[ibrahiemy]

i just want to know how to use the GPU instead of CPU
i tried on rtx 2060 and rtx3080ti
but it uses 0% of it
i used this commands
--aimg imagine "small bird drinking water"
--imagine "small bird drinking water"

none of them used the gpu
im running windows 10

[brycedrennan]

Sounds like the version of torch installed doesn't know how to talk to cuda. Please output the full logs of when you run a command.

[ibrahiemy]

first i would like to thank you for fast response

this is the log
"
PS C:\Windows\system32> aimg imagine "bird on tree"
C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\pydantic_internal_fields.py:149: UserWarning: Field "model_config_path" has conflict with protected namespace "model_".

You may be able to resolve this warning by setting model_config['protected_namespaces'] = ().
warnings.warn(
✨ Generate images faster using a persistent shell session. Just run aimg to start. This makes generation and editing much quicker since the model can stay loaded in memory.

Received 1 prompt(s) and 0 input image(s). Will repeat the generations 1 times to create 1 images.
Running in CPU mode. it's gonna be slooooooow.
🖼 Generating 1/1: "bird on tree" 512x512px negative-prompt:"Ugly, duplication, duplicates, mutila tion, deformed, mutilated, mutation, twisted body, disfigured, bad anatomy, out of frame, extra fingers, mutated hands, poorly drawn hands, extra limbs, malformed limbs, missing arms, extra arms, missing legs, extra legs, mutated hands, extra hands, fused fingers, missing fingers, extra fingers, long neck, small head, closed eyes, rolling eyes, weird eyes, smudged face, blurred face, poorly drawn face, mutation, mutilation, cloned face, strange mouth, grainy, blurred, blurry, writing, calligraphy, signature, text, watermark, bad art," seed:None prompt-strength:7.5 steps:15 sampler-type:k_dpmpp_2m init-image-strength:0.2 model:SD-1.5
Loading model C:\Users\Cruel.cache\huggingface\hub\models--runwayml--stable-diffusion-v1-5\snapshots\889b629140e71758e1e0006e355c331a5744b4bf\v1-5-pruned-emaonly.ckpt onto cpu backend...
100%|██████████████████████████████████████████████████████████████████████████████████| 14/14 [01:22<00:00, 5.92s/it]
Image Generated. Timings: conditioning:0.10s sampling:82.83s decoding:3.74s safety-filter:1.77s total:88.45s
[generated] saved to: ./outputs\generated\000008_354535215_kdpmpp2m15_PS7.5_bird_on_tree_[generated].jpg"

[ibrahiemy]

this is the torch version
PS C:\Windows\system32> pip show torch
Name: torch
Version: 1.13.1
Summary: Tensors and Dynamic neural networks in Python with strong GPU acceleration
Home-page: https://pytorch.org/
Author: PyTorch Team
Author-email: [email protected]
License: BSD-3
Location: c:\users\cruel\appdata\local\programs\python\python310\lib\site-packages
Requires: typing-extensions
Required-by: xformers, torchvision, torchmetrics, torchdiffeq, timm, pytorch-lightning, open-clip-torch, kornia, imaginAIry, fairscale, facexlib

[ibrahiemy]

i tried this command to update torch
pip install --upgrade torch torchvision

then this log started to show

"PS C:\Windows\system32> aimg imagine "bird on tree"
C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\pydantic_internal_fields.py:149: UserWarning: Field "model_config_path" has conflict with protected namespace "model_".

You may be able to resolve this warning by setting model_config['protected_namespaces'] = ().
warnings.warn(
✨ Generate images faster using a persistent shell session. Just run aimg to start. This makes generation and editing much quicker since the model can stay loaded in memory.

Received 1 prompt(s) and 0 input image(s). Will repeat the generations 1 times to create 1 images.
Traceback (most recent call last):
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 196, in _run_module_as_main
return _run_code(code, main_globals, None,
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\runpy.py", line 86, in run_code
exec(code, run_globals)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\Scripts\aimg.exe_main.py", line 7, in
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\core.py", line 1157, in call
return self.main(*args, **kwargs)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\core.py", line 1078, in main
rv = self.invoke(ctx)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click_shell\core.py", line 164, in invoke
ret = super(Shell, self).invoke(ctx)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\core.py", line 1688, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\core.py", line 1434, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\core.py", line 783, in invoke
return __callback(*args, **kwargs)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\click\decorators.py", line 33, in new_func
return f(get_current_context(), *args, **kwargs)
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\imaginairy\cli\imagine.py", line 144, in imagine_cmd
return _imagine_cmd(
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\imaginairy\cli\shared.py", line 181, in imagine_cmd
filenames = imagine_image_files(
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\imaginairy\api.py", line 61, in imagine_image_files
for result in imagine(
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\imaginairy\api.py", line 153, in imagine
check_torch_version()
File "C:\Users\Cruel\AppData\Local\Programs\Python\Python310\lib\site-packages\imaginairy\utils_init.py", line 255, in check_torch_version
raise RuntimeError("ImaginAIry is not compatible with torch>=2.0.0")
RuntimeError: ImaginAIry is not compatible with torch>=2.0.0
PS C:\Windows\system32>"

[ibrahiemy]

i tried to use imaginary
14.0.4
14.0.3
14.0.2
14.0.1
14.0.0
13.2.1
all with all versions of torch 1.13.1 and 2.0.0 and 2.1.2
none worked

[brycedrennan]

I would install torch from here and use the most recent imaginairy
https://pytorch.org/get-started/locally/

You can also run aimg system-info to display more relevant information. If that doesn't help then I'd guess you don't have the right graphics card drivers installed. Hard to say though.

[ibrahiemy]

thanks for the tips
that guided me to find the solution
which was this code
pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu117