Working on large graph dataset - amazon_photo

[Amalsalem]

i have a message while ruining the mode for large graph dataset: the says that i need to install the cuda version of dgl

Notebook : loadamazonphoto.ipynb

Traceback (most recent call last): File "/home/nfs/20015279/Dink-Net-main/main.py", line 184, in train(args=args) File "/home/nfs/20015279/Dink-Net-main/main.py", line 53, in train x, adj, model = map(lambda tmp: tmp.to(args.device), [x, adj, model]) File "/home/nfs/20015279/Dink-Net-main/main.py", line 53, in x, adj, model = map(lambda tmp: tmp.to(args.device), [x, adj, model]) File "/home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/heterograph.py", line 5709, in to ret._graph = self._graph.copy_to(utils.to_dgl_context(device)) File "/home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/heterograph_index.py", line 255, in copy_to return _CAPI_DGLHeteroCopyTo(self, ctx.device_type, ctx.device_id) File "dgl/_ffi/_cython/./function.pxi", line 295, in dgl._ffi._cy3.core.FunctionBase.call File "dgl/_ffi/_cython/./function.pxi", line 227, in dgl._ffi._cy3.core.FuncCall File "dgl/_ffi/_cython/./function.pxi", line 217, in dgl._ffi._cy3.core.FuncCall3 dgl._ffi.base.DGLError: [07:51:53] /opt/dgl/src/runtime/c_runtime_api.cc:82: Check failed: allow_missing: Device API cuda is not enabled. Please install the cuda version of dgl. Stack trace: [bt] (0) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x75) [0x7f44d64f38f5] [bt] (1) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::runtime::DeviceAPIManager::GetAPI(std::string, bool)+0x202) [0x7f44d6862d82] [bt] (2) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::runtime::DeviceAPI::Get(DGLContext, bool)+0x1e1) [0x7f44d685f361] [bt] (3) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::runtime::NDArray::Empty(std::vector<long, std::allocator >, DGLDataType, DGLContext)+0x13b) [0x7f44d687a83b] [bt] (4) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::runtime::NDArray::CopyTo(DGLContext const&) const+0xc3) [0x7f44d68b5043] [bt] (5) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::UnitGraph::CopyTo(std::shared_ptr, DGLContext const&)+0x3ff) [0x7f44d69c253f] [bt] (6) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(dgl::HeteroGraph::CopyTo(std::shared_ptr, DGLContext const&)+0xf6) [0x7f44d68c18c6] [bt] (7) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(+0x51b686) [0x7f44d68d0686] [bt] (8) /home/nfs/20015279/.local/lib/python3.9/site-packages/dgl/libdgl.so(DGLFuncCall+0x48) [0x7f44d685e598]

[asubah]

You can check their start page here https://www.dgl.ai/pages/start.html I recommand using the containers through singularity / apptainer.

[Amalsalem]

Do you mean I need to install the needed in my environment !

Can you prepare for me an environment in the server for this issue?

On Sun, Dec 3, 2023 at 5:36 PM Abdulla Subah @.***> wrote:

You can check their start page here https://www.dgl.ai/pages/start.html I recommand using the containers through singularity / apptainer.

— Reply to this email directly, view it on GitHub https://github.com/UOB-AI/UOB-AI.github.io/issues/48#issuecomment-1837501940, or unsubscribe https://github.com/notifications/unsubscribe-auth/AY3B76AUPVZV5AOXWTL25DTYHSFANAVCNFSM6AAAAAA7626QTKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMZXGUYDCOJUGA . You are receiving this because you authored the thread.Message ID: @.***>

[asubah]

If you are going to use the container you don't need to install anything, you only need to download the container image. In case it is too big, you can put it under /data/dataset as usual.

If you want a Conda environment, please specify which frameworks are you going to use the library with, PyTorch, TensorFlow, or MXNET. And any other packages you need.