Error launching after installing requirements about draggan

[Zeqiang-Lai]

See #14 (comment)

[crazygonuts]

CUDA and visual studio are already installed. I reinstalled anyways, but am still getting the same error. I am on windows 10.

[DuckersMcQuack]

Same here. Worked fine with the old version of this drag gan from a few days ago. new one can't find cuda either. And i have cuda 11 installed, path's setup properly and everything. And works perfectly fine with any other diffusion related image generator.

[werwack]

Hello, I have the same issue on my side too. Ideas?
Thank you

[raiden2012]

Try install torch with cuda support if it doesn't. My cmd is
pip install torch==2.0.1+cu118 --extra-index-url https://download.pytorch.org/whl/cu118

[crazygonuts]

Try install torch with cuda support if it doesn't. My cmd is pip install torch==2.0.1+cu118 --extra-index-url https://download.pytorch.org/whl/cu118

tried this, torch was already installed so it reinstalled but the error is still there

[Anastasia-Zyh]

Hello, I have the same issue on my side too. any ideas?
Thank you

[gonduras]

same shit

[Zeqiang-Lai]

same shit

The latest version would automatically fallback to pytorch native implementation without custom op, which should not need CUDA toolkit anymore.

What the error messages do you face ?

[crazygonuts]

On latest ( 4015f1f ) I am now getting this:

No CUDA runtime is found, using CUDA_HOME='C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1'
D:\AIModels\Art\GAN\DragGAN\venv\lib\site-packages\torch\utils\cpp_extension.py:359: UserWarning: Error checking compiler version for cl: [WinError 2] The system cannot find the file specified
  warnings.warn(f'Error checking compiler version for {compiler}: {error}')
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\fused_act.py:22: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\upfirdn2d.py:21: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
Traceback (most recent call last):
  File "D:\AIModels\Art\GAN\DragGAN\gradio_app.py", line 10, in <module>
    from stylegan2.inversion import inverse_image
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\inversion.py", line 12, in <module>
    from .lpips import util
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\util.py", line 11, in <module>
    from . import dist_model
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\dist_model.py", line 12, in <module>
    from .base_model import BaseModel
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\base_model.py", line 6, in <module>
    from IPython import embed
ModuleNotFoundError: No module named 'IPython'

[Zeqiang-Lai]

On latest ( 4015f1f ) I am now getting this:

No CUDA runtime is found, using CUDA_HOME='C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1'
D:\AIModels\Art\GAN\DragGAN\venv\lib\site-packages\torch\utils\cpp_extension.py:359: UserWarning: Error checking compiler version for cl: [WinError 2] The system cannot find the file specified
  warnings.warn(f'Error checking compiler version for {compiler}: {error}')
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\fused_act.py:22: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\upfirdn2d.py:21: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
Traceback (most recent call last):
  File "D:\AIModels\Art\GAN\DragGAN\gradio_app.py", line 10, in <module>
    from stylegan2.inversion import inverse_image
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\inversion.py", line 12, in <module>
    from .lpips import util
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\util.py", line 11, in <module>
    from . import dist_model
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\dist_model.py", line 12, in <module>
    from .base_model import BaseModel
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\base_model.py", line 6, in <module>
    from IPython import embed
ModuleNotFoundError: No module named 'IPython'

pip install IPython

this should fix it. we will update the package soon

[DuckersMcQuack]

On latest ( 4015f1f ) I am now getting this:

No CUDA runtime is found, using CUDA_HOME='C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1'
D:\AIModels\Art\GAN\DragGAN\venv\lib\site-packages\torch\utils\cpp_extension.py:359: UserWarning: Error checking compiler version for cl: [WinError 2] The system cannot find the file specified
  warnings.warn(f'Error checking compiler version for {compiler}: {error}')
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\fused_act.py:22: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
D:\AIModels\Art\GAN\DragGAN\stylegan2\op\upfirdn2d.py:21: UserWarning: Fail to build torch extension, switch to native implementation
  warnings.warn(
Traceback (most recent call last):
  File "D:\AIModels\Art\GAN\DragGAN\gradio_app.py", line 10, in <module>
    from stylegan2.inversion import inverse_image
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\inversion.py", line 12, in <module>
    from .lpips import util
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\util.py", line 11, in <module>
    from . import dist_model
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\dist_model.py", line 12, in <module>
    from .base_model import BaseModel
  File "D:\AIModels\Art\GAN\DragGAN\stylegan2\lpips\base_model.py", line 6, in <module>
    from IPython import embed
ModuleNotFoundError: No module named 'IPython'

pip install IPython

this should fix it. we will update the package soon

Just did, still the same as the lads above.

[ovanderz]

I'm getting the same error. Looking into the cpp_extension.py I get this;

def get_compiler_abi_compatibility_and_version(compiler) -> Tuple[bool, TorchVersion]:
r'''
Determine if the given compiler is ABI-compatible with PyTorch alongside
its version.

Args:
    compiler (str): The compiler executable name to check (e.g. ``g++``).
        Must be executable in a shell process.

Returns:
    A tuple that contains a boolean that defines if the compiler is (likely) ABI-incompatible with PyTorch,
    followed by a `TorchVersion` string that contains the compiler version separated by dots.
'''
if not _is_binary_build():
    return (True, TorchVersion('0.0.0'))
if os.environ.get('TORCH_DONT_CHECK_COMPILER_ABI') in ['ON', '1', 'YES', 'TRUE', 'Y']:
    return (True, TorchVersion('0.0.0'))

# First check if the compiler is one of the expected ones for the particular platform.
if not check_compiler_ok_for_platform(compiler):
    warnings.warn(WRONG_COMPILER_WARNING.format(
        user_compiler=compiler,
        pytorch_compiler=_accepted_compilers_for_platform()[0],
        platform=sys.platform))
    return (False, TorchVersion('0.0.0'))

if IS_MACOS:
    # There is no particular minimum version we need for clang, so we're good here.
    return (True, TorchVersion('0.0.0'))
try:
    if IS_LINUX:
        minimum_required_version = MINIMUM_GCC_VERSION
        versionstr = subprocess.check_output([compiler, '-dumpfullversion', '-dumpversion'])
        version = versionstr.decode(*SUBPROCESS_DECODE_ARGS).strip().split('.')
    else:
        minimum_required_version = MINIMUM_MSVC_VERSION
        compiler_info = subprocess.check_output(compiler, stderr=subprocess.STDOUT)
        match = re.search(r'(\d+)\.(\d+)\.(\d+)', compiler_info.decode(*SUBPROCESS_DECODE_ARGS).strip())
        version = ['0', '0', '0'] if match is None else list(match.groups())
except Exception:
    _, error, _ = sys.exc_info()
    warnings.warn(f'Error checking compiler version for {compiler}: {error}')
    return (False, TorchVersion('0.0.0'))

if tuple(map(int, version)) >= minimum_required_version:
    return (True, TorchVersion('.'.join(version)))

compiler = f'{compiler} {".".join(version)}'
warnings.warn(ABI_INCOMPATIBILITY_WARNING.format(compiler))

return (False, TorchVersion('.'.join(version)))

It seems to me like there isn't an ABI compatability option for Windows (maybe?) (I'm a noob. does this make any sense?)