Giters

NVIDIA / vid2vid

Pytorch implementation of our method for high-resolution (e.g. 2048x1024) photorealistic video-to-video translation.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

no module named channelnorm_cuda

venkat8900 opened this issue · comments

commented

ran this code on colab : !bash ./scripts/face/train_512.sh
And its showing me the following error. Please help me to resolve this

Traceback (most recent call last):
File "train.py", line 148, in
train()
File "train.py", line 28, in train
models = create_model(opt)
File "/content/drive/My Drive/Colab Notebooks/pix2pix_video_synthesis_NVDIA/vid2vid/models/models.py", line 79, in create_model
flowNet.initialize(opt)
File "/content/drive/My Drive/Colab Notebooks/pix2pix_video_synthesis_NVDIA/vid2vid/models/flownet.py", line 14, in initialize
from .flownet2_pytorch import models as flownet2_models
File "/content/drive/My Drive/Colab Notebooks/pix2pix_video_synthesis_NVDIA/vid2vid/models/flownet2_pytorch/models.py", line 9, in
from .networks.channelnorm_package.channelnorm import ChannelNorm
File "/content/drive/My Drive/Colab Notebooks/pix2pix_video_synthesis_NVDIA/vid2vid/models/flownet2_pytorch/networks/channelnorm_package/channelnorm.py", line 3, in
import channelnorm_cuda
ModuleNotFoundError: No module named 'channelnorm_cuda'

commented 
cd models/networks/flownet2_pytorch/networks/channelnorm_package

python setup.py build
python setup.py install
commented 
cd models/networks/flownet2_pytorch/networks/channelnorm_package

python setup.py build
python setup.py install

I tried as you suggested, but still the same error is coming. I am trying to implement, few shot vid2vid in google colab. But, the same error is coming.