compilation fails on example 05

[MDCurrent]

feel free to close if this is not constructive but I haven't been able to succesfully run this code for a couple of weeks

# create image
./docker/build.sh cuda
# run image 
docker run --gpus all -it ait:latest
# download pipeline 
python3 scripts/download_pipeline.py --token ...

# compile SD 
python3 scripts/compile.py
 RuntimeError: Failed to create cutlass library lib: [Errno 2] No such file or directory: '/usr/local/lib/python3.8/3rdparty/cutlass/tools/library/scripts'

# check that dir 
ls /usr/local/lib/python3.8/
dist-packages

I assume im overlooking something dead simple but wanted to see if you all might know what is causing this. I've been keeping up to date with main as I've struggled through this

[tenpercent]

Hi @MDCurrent - just in case, have you cloned the submodules under 3rdparty folder before creating a Docker image?

[MDCurrent]

I see the 3rdparty modules on my host machine in the AITemplate repo yea

/AITemplate$ ls 3rdparty/
composable_kernel  cub  cutlass  picojson

but I haven't manually cloned them to the usr/local/lib/python3.8/ directory

Thank you for your time and attention

[MDCurrent]

nvidia-smi --query-gpu=compute_cap --format=csv
compute_cap
7.5

Looks like my card is not the proper generation. Will close this to represent the hardware incompatibility