Fail to run with CUDA 11.3

[TB5z035]

Hi! This is a really exciting work, and thanks for your effort in this implementation.

Unfortunately, I cannot recreate the results on my own server equipped with a NVIDIA A100 (which would require >= CUDA 11.1). As the text2mesh.yml would prepare an unavailable environment with CUDA 10.2, I set up the environment by myself with the following commands:

conda create -n text2mesh python=3.7
conda activate text2mesh
export TORCH_CUDA_ARCH_LIST="8.6"
cd kaolin/
git checkout v0.9.1
python setup.py develop
pip install git+https://github.com/openai/CLIP.git
pip install matplotlib

However, when I follow the instruction and run ./demo/run_vase.sh, I cannot observe any non-trivial results. Each 100 iterations would generate the same intermediate results:

and vase_final.obj doesn't show any styled details.

The output of the codebase:

Is there any workaround? Thanks!

[factoryofthesun]

Hi thanks for trying out the code! I've been able to successfully run the scripts with CUDA 11.1, python 3.8, and pytorch 1.9 on A40s, but I don't think we've tested on A100s specifically. Note that kaolin requires pytorch <= 1.9, so it shouldn't have been able to install properly on CUDA > 11.1. Can you try creating an environment with the above versions and see if it works?