search

autonomousvision / stylegan-xl

[SIGGRAPH'22] StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets
MIT License
965 stars 112 forks source link

AttributeError #56

Closed Limbicnation closed 2 years ago

Limbicnation commented 2 years ago

When I try to generated videos I get the following error:

    raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'DummyMapping' object has no attribute 'w_avg

I appreciate your help!

woctezuma commented 2 years ago

What is the netkwork which you load?

https://github.com/autonomousvision/stylegan_xl/blob/f9be58e98110bd946fcdadef2aac8345466faaf3/gen_video.py#L185-L188

Limbicnation commented 2 years ago

Hi thanks for the response! I am loading the network-snapshot.pkl in training-runs/00001-fastgan-tumblr256-gpus1-batch64

woctezuma commented 2 years ago

Does this command work?

python gen_video.py --output=lerp.mp4 --trunc=0.7 --seeds=0-31 --grid=4x2 \
  --network=https://s3.eu-central-1.amazonaws.com/avg-projects/stylegan_xl/models/pokemon256.pkl

Reference: https://github.com/autonomousvision/stylegan_xl#generating-samples--interpolations

If the answer is yes, then there is an issue with the .pkl at training-runs/00001-fastgan-tumblr256-gpus1-batch64.

If the answer is no, then there is an issue with the code.

Limbicnation commented 2 years ago

Hi, thanks for the response! When I run the above code, I get the following error:

nvcc fatal : Unsupported gpu architecture 'compute_86' [2/3] c++ -MMD -MF bias_act.o.d -DTORCH_EXTENSION_NAME=bias_act_plugin -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /home/ws-ml/anaconda3/envs/sgxl/lib/python3.9/site-packages/torch/include -isystem /home/ws-ml/anaconda3/envs/sgxl/lib/python3.9/site-packages/torch/include/torch/csrc/api/include -isystem /home/ws-ml/anaconda3/envs/sgxl/lib/python3.9/site-packages/torch/include/TH -isystem /home/ws-ml/anaconda3/envs/sgxl/lib/python3.9/site-packages/torch/include/THC -isystem /home/ws-ml/anaconda3/envs/sgxl/include/python3.9 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -c /home/ws-ml/.cache/torch_extensions/py39_cu111/bias_act_plugin/3cb576a0039689487cfba59279dd6d46-nvidia-geforce-rtx-3080-ti/bias_act.cpp -o bias_act.o ninja: build stopped: subcommand failed

I am guessing that I am not running the correct Cuda Toolkit ?

image

xl-sr commented 2 years ago

this seems like a problem related to the custom cuda kernels. Are you able to run the original StyleGAN 3 repo?

Limbicnation commented 2 years ago

@xl-sr I have not tried to run the original StyleGAN 3 repo yet. I might have to do that first. Thanks!