search

DeepGraphLearning / NBFNet

Official implementation of Neural Bellman-Ford Networks (NeurIPS 2021)
MIT License
197 stars 29 forks source link

Problems about ninja #10

Open Robot-2020 opened 2 years ago

Robot-2020 commented 2 years ago

Hi, Doctor. I meet some problems when I run the code on the Linux. I do really need your help. Could you help me? It really troubles me a lot.

15:43:32   Preprocess training set
15:43:36   >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
15:43:36   Epoch 0 begin
Traceback (most recent call last):
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1666, in _run_ninja_build
    subprocess.run(
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/subprocess.py", line 516, in run
    raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "script/run.py", line 62, in <module>
    train_and_validate(cfg, solver)
  File "script/run.py", line 27, in train_and_validate
    solver.train(**kwargs)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/core/engine.py", line 143, in train
    loss, metric = model(batch)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/tasks/reasoning.py", line 85, in forward
    pred = self.predict(batch, all_loss, metric)
  File "/data1/home/wza/nbfnet/nbfnet/task.py", line 288, in predict
    pred = self.model(graph, h_index, t_index, r_index, all_loss=all_loss, metric=metric)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data1/home/wza/nbfnet/nbfnet/model.py", line 149, in forward
    output = self.bellmanford(graph, h_index[:, 0], r_index[:, 0])
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/decorator.py", line 232, in fun
    return caller(func, *(extras + args), **kw)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/utils/decorator.py", line 56, in wrapper
    return forward(self, *args, **kwargs)
  File "/data1/home/wza/nbfnet/nbfnet/model.py", line 115, in bellmanford
    hidden = layer(step_graph, layer_input)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
    return forward_call(*input, **kwargs)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/conv.py", line 91, in forward
    update = self.message_and_aggregate(graph, input)
  File "/data1/home/wza/nbfnet/nbfnet/layer.py", line 140, in message_and_aggregate
    sum = functional.generalized_rspmm(adjacency, relation_input, input, sum="add", mul=mul)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/spmm.py", line 378, in generalized_rspmm
    return Function.apply(sparse.coalesce(), relation, input)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/spmm.py", line 172, in forward
    forward = spmm.rspmm_add_mul_forward_cuda
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/utils/torch.py", line 27, in __getattr__
    return getattr(self.module, key)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/utils/decorator.py", line 21, in __get__
    result = self.func(obj)
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/utils/torch.py", line 31, in module
    return cpp_extension.load(self.name, self.sources, self.extra_cflags, self.extra_cuda_cflags,
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1080, in load
    return _jit_compile(
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1293, in _jit_compile
    _write_ninja_file_and_build_library(
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1405, in _write_ninja_file_and_build_library
    _run_ninja_build(
  File "/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1682, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'spmm': [1/3] /usr/local/cuda-10.2/bin/nvcc  -DTORCH_EXTENSION_NAME=spmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/TH -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda-10.2/include -isystem /data1/home/wza/.conda/envs/linkp/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_70,code=compute_70 -gencode=arch=compute_70,code=sm_70 --compiler-options '-fPIC' -O3 -std=c++14 -c /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu -o rspmm.cuda.o
FAILED: rspmm.cuda.o
/usr/local/cuda-10.2/bin/nvcc  -DTORCH_EXTENSION_NAME=spmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/TH -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda-10.2/include -isystem /data1/home/wza/.conda/envs/linkp/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_70,code=compute_70 -gencode=arch=compute_70,code=sm_70 --compiler-options '-fPIC' -O3 -std=c++14 -c /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu -o rspmm.cuda.o
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu: In instantiation of ‘at::rspmm_forward_cuda(const SparseTensor&, const at::Tensor&, const at::Tensor&)::<lambda()>::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]’:
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:246:600:   required from ‘struct at::rspmm_forward_cuda(const SparseTensor&, const at::Tensor&, const at::Tensor&)::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]::<lambda()>’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:246:608:   required from ‘at::rspmm_forward_cuda(const SparseTensor&, const at::Tensor&, const at::Tensor&)::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:246:607:   required from ‘struct at::rspmm_forward_cuda(const SparseTensor&, const at::Tensor&, const at::Tensor&) [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul; at::sparse::SparseTensor = at::Tensor]::<lambda()>’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:246:28:   required from ‘at::Tensor at::rspmm_forward_cuda(const SparseTensor&, const at::Tensor&, const at::Tensor&) [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul; at::sparse::SparseTensor = at::Tensor]’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:356:193:   required from here
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/rspmm.cu:244:37: internal compiler error: in tsubst_copy, at cp/pt.c:13189
     const int num_row_block = (num_row + row_per_block - 1) / row_per_block;
                                     ^
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-5/README.Bugs> for instructions.
[2/3] /usr/local/cuda-10.2/bin/nvcc  -DTORCH_EXTENSION_NAME=spmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/TH -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda-10.2/include -isystem /data1/home/wza/.conda/envs/linkp/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_70,code=compute_70 -gencode=arch=compute_70,code=sm_70 --compiler-options '-fPIC' -O3 -std=c++14 -c /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu -o spmm.cuda.o
FAILED: spmm.cuda.o
/usr/local/cuda-10.2/bin/nvcc  -DTORCH_EXTENSION_NAME=spmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/TH -isystem /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torch/include/THC -isystem /usr/local/cuda-10.2/include -isystem /data1/home/wza/.conda/envs/linkp/include/python3.8 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_70,code=compute_70 -gencode=arch=compute_70,code=sm_70 --compiler-options '-fPIC' -O3 -std=c++14 -c /data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu -o spmm.cuda.o
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu: In instantiation of ‘at::spmm_forward_cuda(const SparseTensor&, const at::Tensor&)::<lambda()>::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]’:
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:219:506:   required from ‘struct at::spmm_forward_cuda(const SparseTensor&, const at::Tensor&)::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]::<lambda()>’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:219:514:   required from ‘at::spmm_forward_cuda(const SparseTensor&, const at::Tensor&)::<lambda()> [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul]’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:219:512:   required from ‘struct at::spmm_forward_cuda(const SparseTensor&, const at::Tensor&) [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul; at::sparse::SparseTensor = at::Tensor]::<lambda()>’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:219:28:   required from ‘at::Tensor at::spmm_forward_cuda(const SparseTensor&, const at::Tensor&) [with NaryOp = at::NaryAdd; BinaryOp = at::BinaryMul; at::sparse::SparseTensor = at::Tensor]’
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:315:157:   required from here
/data1/home/wza/.conda/envs/linkp/lib/python3.8/site-packages/torchdrug/layers/functional/extension/spmm.cu:217:37: internal compiler error: in tsubst_copy, at cp/pt.c:13189
     const int num_row_block = (num_row + row_per_block - 1) / row_per_block;
                                     ^
Please submit a full bug report,
with preprocessed source if appropriate.
See <file:///usr/share/doc/gcc-5/README.Bugs> for instructions.
ninja: build stopped: subcommand failed.
Robot-2020 commented 2 years ago

When I evaluate ogbl-biokg by cpu There is an error saying that: RuntimeError: [enforce fail at ..\c10\core\CPUAllocator.cpp:75] data. DefaultCPUAllocator: not enough memory: you tried to allocate 3917082528 bytes. Buy new RAM!

How can I resolve it? Please!

KiddoZhu commented 2 years ago

Hi! I haven't encountered the ninja issue you post here. It looks like ninja reports some internal compiler error?

For the memory error, it tries to allocate around 3GB memory. I think that's a expected behavior for large datasets like ogbl-biokg. Maybe you don't have enough memory in your machine?

Robot-2020 commented 2 years ago

I'm thinking about whether the ninja issue was caused by the gcc version? Unfortunately I'm just a normal student in Chineses college, I can only borrow the school's server. No permission to update gcc. So, when I change to run the code on the Windows, I can successfully trained the model. But in the test result, the report memory is not enough error. I have checked it, it still has about 10GB memory, so, I don't know why it reports this error. By the way, How can I do it so I can only test a part to get the result? Thank you so much.

chandeler commented 1 year ago

I also met something wrong about ninja on a windows10 +python3.7+torch1.8 machine

nxznm commented 1 year ago

@Robot-2020 Hi~ Recently, I met the same issue as yours. I think the problem is caused by the gcc version. My original gcc version in my lab server machine is 5.3. Now I use anaconda to install gcc & g++ (their version is 7.3). And now there is no such issue:) Maybe you can refer to this website.