uw-ipd/RoseTTAFold2NA

Cannot load Graphbolt C++ library

SuhasSrinivasan opened this issue · 2 comments

SuhasSrinivasan commented

Dear Researchers,

Thank you for making this method!

Currently there is a blocking issue when trying to reproduce the example.
Originating from this below step.

python $PIPEDIR/network/predict.py \
    -inputs $argstring \
    -prefix $WDIR/models/model \
    -model $PIPEDIR/network/weights/RF2NA_apr23.pt \
    -db $HHDB #2> $WDIR/log/network.stderr #1> $WDIR/log/network.stdout

There are unmet dependencies when executing prediction.
The straightforward ones are:

  1. torchdata
  2. pandas
  3. pydantic

But after solving these dependencies, there is an issue with DGL

  File "/miniconda3/envs/rf2na/lib/python3.10/site-packages/dgl/graphbolt/__init__.py", line 52, in load_graphbolt
    raise ImportError("Cannot load Graphbolt C++ library")
ImportError: Cannot load Graphbolt C++ library

There have been long standing discussions about unmet dependencies #36, can we please get a solution for this.

SuhasSrinivasan commented

Noting Nvidia and PyTorch packages

# Name                    Version                   Build  Channel

cpuonly                   2.0                           0    pytorch

cuda-cudart               11.7.99                       0    nvidia
cuda-cupti                11.7.101                      0    nvidia
cuda-libraries            11.7.1                        0    nvidia
cuda-nvrtc                11.7.99                       0    nvidia
cuda-nvtx                 11.7.91                       0    nvidia
cuda-runtime              11.7.1                        0    nvidia

dgl                       2.1.0.cu117             py310_0    dglteam/label/cu117
dllogger                  1.0.0                    pypi_0    pypi

libcublas                 11.10.3.66                    0    nvidia
libcufft                  10.7.2.124           h4fbf590_0    nvidia
libcufile                 1.9.0.20                      0    nvidia
libcurand                 10.3.5.119                    0    nvidia
libcusolver               11.4.0.1                      0    nvidia
libcusparse               11.7.4.91                     0    nvidia

libnpp                    11.7.4.75                     0    nvidia
libnvjpeg                 11.8.0.2                      0    nvidia

pytorch                   2.0.1              py3.10_cpu_0    pytorch
pytorch-cuda              11.7                 h778d358_5    pytorch
pytorch-mutex             1.0                         cpu    pytorch
SuhasSrinivasan commented

Solution here #105