Open M202071895 opened 3 years ago
Hi there, I would recommend you completely remove the previously built modules, for instance, ~/.local/python/xxx/softrasxxx.egg. and then rebuild the repo
my code environment is: ubuntu+pytorch1.1.0+cuda10.0. I can install and run 0th(rendering) successfully, but when I run this: CUDA_VISIBLE_DEVICES=0 python examples/recon/train.py -eid recon, I meet this problem. Anybody else meets the same error on ubuntu?
have u sloved this question? i also meet it.And i tried many times,but failed.
Seemingly, it should be fixed by changing the working directory to make sure the base folder of this repo is not in your sys.path. Otherwise, Python tends to search this soft_renderer.cuda.load_textures in the folder of this repo in higher priority than the packages installed in site-packages.
Fyi: I only got rid of this error (on google colab) by '!pip install git+https://github.com/ShichenLiu/SoftRas' and cloning the repo afterwards to access /data
Fyi: I only got rid of this error (on google colab) by '!pip install git+https://github.com/ShichenLiu/SoftRas' and cloning the repo afterwards to access /data
Hi Used that code before and can't run it again now (no matter which environment I'm using). Tried and colab tip and I got: Building wheels for collected packages: soft-renderer Building wheel for soft-renderer (setup.py) ... error ERROR: Failed building wheel for soft-renderer Running setup.py clean for soft-renderer Failed to build soft-renderer Installing collected packages: soft-renderer Running setup.py install for soft-renderer ... error ERROR: Command errored out with exit status 1: /usr/bin/python3 -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-req-build-btndn4ww/setup.py'"'"'; file='"'"'/tmp/pip-req-build-btndn4ww/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' install --record /tmp/pip-record-74tyc7rn/install-record.txt --single-version-externally-managed --compile --install-headers /usr/local/include/python3.7/soft-renderer Check the logs for full command output.
Any ideas? Thanks
The only way I can reproduce this error in colab is when I'm not connected to GPU. Make sure your runtime is set to GPU. Bests
Thanks, bmiserez - got a few more questions running the code, and I would be happy to contact you by mail if I can - mine is eyal.friedman1977@gmail.com thanks
I wonder if it has anything to do with the CUDA version. When I use the cuda 10.2 version , and the python version is 3.8. After I progress the command " python setup.py install ", run the command"CUDA_VISIBLE_DEVICES=0 python examples/demo_render.py" appear the error : Module Not Found Error: No module named 'soft_renderer.cuda.load'
Seemingly, it should be fixed by changing the working directory to make sure the base folder of this repo is not in your sys.path. Otherwise, Python tends to search this soft_renderer.cuda.load_textures in the folder of this repo in higher priority than the packages installed in site-packages.
Sorry ! I can't understand this completely. Does this mean that this module is not written by the author program but by a third party package? right? If so,what is its name? It's not in the setup.py file , is it? Looking forward to your reply. Many thanks.
my code environment is: ubuntu+pytorch1.1.0+cuda10.0. I can install and run 0th(rendering) successfully, but when I run this: CUDA_VISIBLE_DEVICES=0 python examples/recon/train.py -eid recon, I meet this problem. Anybody else meets the same error on ubuntu?