convert .pkl error that 'utf-8' codec cant't decode

[yangli-lab]

by running the command

(torch_gpu) E:\fork\fork file\stylegan2-pytorch>python convert_weight.py --repo ~/stylegan2 stylegan2-ffhq-config-f.pkl
I got the error that 'utf-8' codec can't decode xxxx: invaliad continuation byte

and the following are the trackback:
D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py:190: UserWarning: Error checking compiler version for cl: 'utf-8' codec can't decode byte 0xd3 in position 0: invalid continuation byte
warnings.warn('Error checking compiler version for {}: {}'.format(compiler, error))
Traceback (most recent call last):
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 1030, in build_extension_module
check=True)
File "D:\Anaconda\envs\torch_gpu\lib\subprocess.py", line 512, in run
output=stdout, stderr=stderr)
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "convert_weight.py", line 11, in
from model import Generator, Discriminator
File "E:\fork\fork file\stylegan2-pytorch\model.py", line 11, in
from op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d, conv2d_gradfix
File "E:\fork\fork file\stylegan2-pytorch\op_init.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "E:\fork\fork file\stylegan2-pytorch\op\fused_act.py", line 15, in
os.path.join(module_path, "fused_bias_act_kernel.cu"),
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 661, in load
is_python_module)
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 830, in _jit_compile
with_cuda=with_cuda)
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 883, in _write_ninja_file_and_build
_build_extension_module(name, build_directory, verbose)
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 1042, in _build_extension_module
message += ": {}".format(error.output.decode())
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xd3 in position 1167: invalid continuation byte

[rosinality]

I think you need to check you can compile custom operators. I don't know much about compiling cuda kernels on windows, but I think you will need to setup cuda build environments and install ninja.

[yangli-lab]

thanks @rosinality, and I've solved the question by replacing the decode setting in file cpp_xetention.py
however, another problem come out.
I used the same command->python convert_weight.py --repo ~/stylegan2 stylegan2-ffhq-config-f.pkl,but an error of no module named fused occured. The following are the details
Traceback (most recent call last):
File "convert_weight.py", line 11, in
from model import Generator, Discriminator
File "E:\fork\fork file\stylegan2-pytorch\model.py", line 11, in
from op import FusedLeakyReLU, fused_leaky_relu, upfirdn2d, conv2d_gradfix
File "E:\fork\fork file\stylegan2-pytorch\op_init_.py", line 1, in
from .fused_act import FusedLeakyReLU, fused_leaky_relu
File "E:\fork\fork file\stylegan2-pytorch\op\fused_act.py", line 15, in
os.path.join(module_path, "fused_bias_act_kernel.cu"),
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 663, in load
is_python_module)
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 843, in _jit_compile
return _import_module_from_library(name, build_directory, is_python_module)
File "D:\Anaconda\envs\torch_gpu\lib\site-packages\torch\utils\cpp_extension.py", line 1050, in _import_module_from_library
file, path, description = imp.find_module(module_name, [path])
File "D:\Anaconda\envs\torch_gpu\lib\imp.py", line 296, in find_module
raise ImportError(_ERR_MSG.format(name), name=name)
ImportError: No module named 'fused'
do you know the reason? my thanks

[rosinality]

Have you tested you can build custom operations? I think the problem is at it. (It is suggested by that subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1. occurred in the error message.)

[yangli-lab]

sorry for the late reply, I've fixed this problem by using the method provided here
Once again, thanks for your patience and the helpful work you've done

[yangli-lab]

sorry for the late reply, I've fixed this problem by using the method provided here
Once again, thanks for your patience and the helpful work you've done