Hi, recently i have met a problem. When i try to use dgl-cu110 and pytorch version 1.10.2 to calculate graph learning, there raise a issue like:
Traceback (most recent call last):
File “D:\Postgraduate\GCN4KGC-main\WGCN\run.py”, line 503, in
main(args)
File “D:\Postgraduate\GCN4KGC-main\WGCN\run.py”, line 412, in main
pred = model.forward(g, all_rel, e1_batch, rel_batch, entity_id)
File “D:\Postgraduate\GCN4KGC-main\WGCN\run.py”, line 219, in forward
x = self.gc2(g, all_edge, emb_initial)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\torch\nn\modules\module.py”, line 889, in _call_impl
result = self.forward(*input, **kwargs)
File “D:\Postgraduate\GCN4KGC-main\WGCN\run.py”, line 70, in forward
g.update_all(fn.u_mul_e(‘ft’, ‘a’, ‘m’), fn.sum(‘m’, ‘ft’))
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\heterograph.py”, line 4849, in update_all
ndata = core.message_passing(g, message_func, reduce_func, apply_node_func)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\core.py”, line 322, in message_passing
ndata = invoke_gspmm(g, mfunc, rfunc)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\core.py”, line 288, in invoke_gspmm
z = op(graph, x, y)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\ops\spmm.py”, line 149, in func
return gspmm(g, binary_op, reduce_op, x, y)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\ops\spmm.py”, line 77, in gspmm
lhs_data, rhs_data)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\backend\pytorch\sparse.py”, line 425, in gspmm
return GSpMM.apply(gidx, op, reduce_op, lhs_data, rhs_data)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\torch\cuda\amp\autocast_mode.py”, line 217, in decorate_fwd
return fwd(*args, **kwargs)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\backend\pytorch\sparse.py”, line 90, in forward
out, (argX, argY) = _gspmm(gidx, op, reduce_op, X, Y)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl\sparse.py”, line 162, in _gspmm
arg_e_nd)
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl_ffi_ctypes\function.py”, line 190, in call
ctypes.byref(ret_val), ctypes.byref(ret_tcode)))
File “D:\Code\Anaconda\envs\GCN4KGC\lib\site-packages\dgl_ffi\base.py”, line 64, in check_call
raise DGLError(py_str(_LIB.DGLGetLastError()))
dgl._ffi.base.DGLError: [10:31:59] C:/Users/Administrator/dgl-0.5/src/array/cuda/spmm.cu:213: Check failed: e == CUSPARSE_STATUS_SUCCESS: CUSPARSE ERROR: 10
- DGL Version:0.7.0
- Backend Library & Version : Pytorch 1.8.2
- OS :Windows
- How you installed DGL :conda
- Python version:3.7
- CUDA/cuDNN version :11.1
- GPU models and configuration：NVIDIA 4060