No module named 'torch' after 4.1 upgrade

Is it expected that the worker environment is using the system gcc/g++ instead of the conda version?

I am running into /usr/bin/gcc and cuda/nvcc from conda potential conflicts on ubuntu 20.04. I have remove the possible conflicting packages from ubuntu (apt remove nvidia-cuda-toolkit…) but ./bin/cryosparcw install-3dflex keeps failing.

Then trying to revert with:
cryoem@myrdal:~/cryosparc2/cryosparc_worker$ ./bin/cryosparcw forcedeps
yields:

...
    gcc -pthread -B /home/cryoem/cryosparc2/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/compiler_compat -Wl,--sysroot=/ -Wsign-compare -fwrapv -Wall -O3 -DNDEBUG -fPIC -DBOOST_ALL_NO_LIB=1 -DBOOST_THREAD_BUILD_DLL=1 -DBOOST_MULTI_INDEX_DISABLE_SERIALIZATION=1 -DBOOST_PYTHON_SOURCE=1 -Dboost=pycudaboost -DBOOST_THREAD_DONT_USE_CHRONO=1 -DPYGPU_PACKAGE=pycuda -DPYGPU_PYCUDA=1 -DHAVE_CURAND=1 -Isrc/cpp -Ibpl-subset/bpl_subset -I/home/cryoem/cryosparc2/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/lib/python3.7/site-packages/numpy/core/include -I/home/cryoem/cryosparc2/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.7m -c src/cpp/cuda.cpp -o build/temp.linux-x86_64-3.7/src/cpp/cuda.o
    In file included from src/cpp/cuda.cpp:4:
    src/cpp/cuda.hpp:14:10: fatal error: cuda.h: No such file or directory
       14 | #include <cuda.h>
          |          ^~~~~~~~
    compilation terminated.
    error: command 'gcc' failed with exit status 1
    ----------------------------------------
ERROR: Command errored out with exit status 1: /home/cryoem/cryosparc2/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/bin/python3.7 -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-req-build-3334l4mw/setup.py'"'"'; __file__='"'"'/tmp/pip-req-build-3334l4mw/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-8lgu7bn9/install-record.txt --single-version-externally-managed --compile --install-headers /home/cryoem/cryosparc2/cryosparc_worker/deps/anaconda/envs/cryosparc_worker_env/include/python3.7m/pycuda Check the logs for full command output.
check_install_deps.sh: 59: ERROR: installing python failed.

I had to re-add nvidia-cuda-toolkit and the system provided cuda10.1

maybe relates to this thread 3DFlex Dependencies; Building pycuda - #10 by qitsweauca

1 Like