kiddozhu / nbfnet-pyg Goto Github PK

Traceback (most recent call last):
  File "script/run.py", line 216, in <module>
    train_and_validate(cfg, model, train_data, valid_data, filtered_data=filtered_data)
  File "script/run.py", line 57, in train_and_validate
    pred = parallel_model(train_data, batch)
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/sk/project/NBFNet-PyG/nbfnet/models.py", line 139, in forward
    output = self.bellmanford(data, h_index[:, 0], r_index[:, 0])  # (num_nodes, batch_size, feature_dim）
  File "/home/sk/project/NBFNet-PyG/nbfnet/models.py", line 104, in bellmanford
    hidden = layer(layer_input, query, boundary, data.edge_index, data.edge_type, size, edge_weight)
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/sk/project/NBFNet-PyG/nbfnet/layers.py", line 66, in forward
    edge_type=edge_type, size=size, edge_weight=edge_weight)
  File "/home/sk/project/NBFNet-PyG/nbfnet/layers.py", line 89, in propagate
    out = self.message_and_aggregate(edge_index, **msg_aggr_kwargs)
  File "/home/sk/project/NBFNet-PyG/nbfnet/layers.py", line 157, in message_and_aggregate
    from nbfnet.rspmm import generalized_rspmm
  File "/home/sk/project/NBFNet-PyG/nbfnet/rspmm/__init__.py", line 1, in <module>
    from .rspmm import generalized_rspmm
  File "/home/sk/project/NBFNet-PyG/nbfnet/rspmm/rspmm.py", line 205, in <module>
    rspmm = load_extension("rspmm", [os.path.join(path, "rspmm.cpp"), os.path.join(path, "rspmm.cu")])
  File "/home/sk/project/NBFNet-PyG/nbfnet/rspmm/rspmm.py", line 199, in load_extension
    return cpp_extension.load(name, sources, extra_cflags, extra_cuda_cflags, **kwargs)
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1091, in load
    keep_intermediates=keep_intermediates)
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1317, in _jit_compile
    return _import_module_from_library(name, build_directory, is_python_module)
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1699, in _import_module_from_library
    file, path, description = imp.find_module(module_name, [path])
  File "/root/anaconda3/envs/NBF_PYG_ENV/lib/python3.7/imp.py", line 296, in find_module
    raise ImportError(_ERR_MSG.format(name), name=name)
ImportError: No module named 'rspmm'

Why？Can you tell me？

Hello,

I was wondering whether there is a particular reason for adding the inverse relations in the graph.

Thanks in advance for your answer!

hi, this repo is a nice job.

But, I can not find a method to make the NBFNet to predict. For example, given a (head, relation) and get a list candidates(tails).

Hi,

Unlike the NBFNet based on TorchDrug, I don't see codes and configs for homogeneous graphs in the PyG version. Is there a way to run homogeneous graphs on the PyG version?

Thanks in advance.

Hi, thank you for your codes! However, I find that the rspmm seems to behave differently from the spmm in TorchDrug.

How I found this:
I tried to load the checkpoint trained by the official repo to the model in this repo and found that the performance was extremely low. I compared the results in each layer and found that in the function message_and_aggregate of the GeneralizedRelationalConv, the result produced by generalized_rspmm is different from the output produced by functional.generalized_rspmm in the official repo, although I have checked that the input is the same.

How to reproduce
In the function message_and_aggregate of the GeneralizedRelationalConv, before the line if self.message_func in self.message2mul:, add:

torch_drug_edge_index = torch.cat([edge_index, edge_type.unsqueeze(0)], dim=0)
torch_drug_edge_weight = edge_weight
adjacency = sparse_coo_tensor(torch_drug_edge_index, torch_drug_edge_weight, size=[num_node, num_node, self.num_relation])
# the adjacency above is the graph.adjacency in the official NBFNet repo, which is produced by https://github.com/DeepGraphLearning/torchdrug/blob/6066fbd82360abb5f270cba1eca560af01b8cc90/torchdrug/data/graph.py#L801
adjacency = adjacency.transpose(0, 1)

and replace the generalized_rspmm below with functional.generalized_rspmm from Torchdrug， just as official repo did (official repo use relation_input, with is the relation in this repo).

also, add necessary import to layers.py.

from torchdrug.layers import functional
from torchdrug.utils import sparse_coo_tensor

I run the commond:

python script/run.py -c config/inductive/wn18rr.yaml --gpus [0] --version v1

and encounted the error, details is:

Traceback (most recent call last):
File "script/run.py", line 216, in
train_and_validate(cfg, model, train_data, valid_data, filtered_data=filtered_data)
File "script/run.py", line 57, in train_and_validate
pred = parallel_model(train_data, batch)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/NBFNet-PyG/nbfnet/models.py", line 139, in forward
output = self.bellmanford(data, h_index[:, 0], r_index[:, 0]) # (num_nodes, batch_size, feature_dim）
File "/root/NBFNet-PyG/nbfnet/models.py", line 104, in bellmanford
hidden = layer(layer_input, query, boundary, data.edge_index, data.edge_type, size, edge_weight)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1110, in _call_impl
return forward_call(*input, **kwargs)
File "/root/NBFNet-PyG/nbfnet/layers.py", line 65, in forward
output = self.propagate(input=input, relation=relation, boundary=boundary, edge_index=edge_index,
File "/root/NBFNet-PyG/nbfnet/layers.py", line 80, in propagate
size = self.check_input(edge_index, size)
File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1185, in getattr
raise AttributeError("'{}' object has no attribute '{}'".format(
AttributeError: 'GeneralizedRelationalConv' object has no attribute 'check_input'

When I run the code, I have the following problem, could you tell me why:

11:21:06 IndWN18RR dataset
11:21:06 #train: 5410, #valid: 630, #test: 188
11:21:08 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
11:21:08 Epoch 0 begin
Load rspmm extension. This may take a while...
Traceback (most recent call last):
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 1673, in _run_ninja_build
env=env)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/subprocess.py", line 418, in run
output=stdout, stderr=stderr)
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 216, in
train_and_validate(cfg, model, train_data, valid_data, filtered_data=filtered_data)
File "script/run.py", line 57, in train_and_validate
pred = parallel_model(train_data, batch)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/nn/modules/module.py", line 889, in call_impl
result = self.forward(*input, **kwargs)
File "/ssd2T/NBFNet-PyG-master/nbfnet/models.py", line 139, in forward
output = self.bellmanford(data, h_index[:, 0], r_index[:, 0]) # (num_nodes, batch_size, feature_dim）
File "/ssd2T/NBFNet-PyG-master/nbfnet/models.py", line 104, in bellmanford
hidden = layer(layer_input, query, boundary, data.edge_index, data.edge_type, size, edge_weight)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/nn/modules/module.py", line 889, in call_impl
result = self.forward(*input, **kwargs)
File "/ssd2T/NBFNet-PyG-master/nbfnet/layers.py", line 66, in forward
edge_type=edge_type, size=size, edge_weight=edge_weight)
File "/ssd2T/NBFNet-PyG-master/nbfnet/layers.py", line 89, in propagate
out = self.message_and_aggregate(edge_index, **msg_aggr_kwargs)
File "/ssd2T/NBFNet-PyG-master/nbfnet/layers.py", line 157, in message_and_aggregate
from .rspmm import generalized_rspmm
File "/ssd2T/NBFNet-PyG-master/nbfnet/rspmm/init.py", line 1, in
from .rspmm import generalized_rspmm
File "/ssd2T/NBFNet-PyG-master/nbfnet/rspmm/rspmm.py", line 204, in
rspmm = load_extension("rspmm", [os.path.join(path, "rspmm.cpp"), os.path.join(path, "rspmm.cu")])
File "/ssd2T/NBFNet-PyG-master/nbfnet/rspmm/rspmm.py", line 199, in load_extension
return cpp_extension.load(name, sources, extra_cflags, extra_cuda_cflags, **kwargs)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 1091, in load
keep_intermediates=keep_intermediates)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 1302, in jit_compile
is_standalone=is_standalone)
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 1407, in write_ninja_file_and_build_library
error_prefix=f"Error building extension '{name}'")
File "/home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/utils/cpp_extension.py", line 1683, in run_ninja_build
raise RuntimeError(message) from e
RuntimeError: Error building extension 'rspmm': [1/2] /usr/bin/nvcc --generate-dependencies-with-compile --dependency-output rspmm.cuda.o.d -DTORCH_EXTENSION_NAME=rspmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/TH -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/THC -isystem /home/czw/anaconda3/envs/NBFNET/include/python3.6m -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_75,code=compute_75 -gencode=arch=compute_75,code=sm_75 --compiler-options '-fPIC' -O3 -std=c++14 -c /ssd2T/NBFNet-PyG-master/nbfnet/rspmm/source/rspmm.cu -o rspmm.cuda.o
FAILED: rspmm.cuda.o
/usr/bin/nvcc --generate-dependencies-with-compile --dependency-output rspmm.cuda.o.d -DTORCH_EXTENSION_NAME=rspmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE="gcc" -DPYBIND11_STDLIB="libstdcpp" -DPYBIND11_BUILD_ABI="cxxabi1011" -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/TH -isystem /home/czw/anaconda3/envs/NBFNET/lib/python3.6/site-packages/torch/include/THC -isystem /home/czw/anaconda3/envs/NBFNET/include/python3.6m -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_75,code=compute_75 -gencode=arch=compute_75,code=sm_75 --compiler-options '-fPIC' -O3 -std=c++14 -c /ssd2T/NBFNet-PyG-master/nbfnet/rspmm/source/rspmm.cu -o rspmm.cuda.o
nvcc fatal : Unknown option '-generate-dependencies-with-compile'
ninja: build stopped: subcommand failed.

12:36:03 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
12:36:03 Epoch 0 begin
Load rspmm extension. This may take a while...
Traceback (most recent call last):
File "script/run.py", line 216, in
train_and_validate(cfg, model, train_data, valid_data, filtered_data=filtered_data)
File "script/run.py", line 57, in train_and_validate
pred = parallel_model(train_data, batch)
File "/home/sln/anaconda3/envs/venv/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/sln/experiment/NBFNet/nbfnet/models.py", line 139, in forward
output = self.bellmanford(data, h_index[:, 0], r_index[:, 0]) # (num_nodes, batch_size, feature_dim）
File "/home/sln/experiment/NBFNet/nbfnet/models.py", line 104, in bellmanford
hidden = layer(layer_input, query, boundary, data.edge_index, data.edge_type, size, edge_weight)
File "/home/sln/anaconda3/envs/venv/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/sln/experiment/NBFNet/nbfnet/layers.py", line 66, in forward
edge_type=edge_type, size=size, edge_weight=edge_weight)
File "/home/sln/experiment/NBFNet/nbfnet/layers.py", line 89, in propagate
out = self.message_and_aggregate(edge_index, **msg_aggr_kwargs)
File "/home/sln/experiment/NBFNet/nbfnet/layers.py", line 182, in message_and_aggregate
sq_sum = generalized_rspmm(edge_index, edge_type, edge_weight, relation ** 2, input ** 2, sum="add",
RuntimeError: CUDA error: no kernel image is available for execution on the device
(venv) [xxx@localhost NBFNet]$ python
Python 3.7.13 (default, Mar 29 2022, 02:18:16)
[GCC 7.5.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.

import torch
torch.cuda.is_available()
True

我参照readme要求配置了环境，但在实际运行中仍然存在cuda报错的问题，但是通过测试torch.cuda.is_available()可知cuda是可用的，根据报错提示看起来似乎是和rspmm相关，请问有什么解决方法吗？

具体环境安装如下：
Package Version

certifi 2022.6.15
charset-normalizer 2.1.0
easydict 1.9
idna 3.3
Jinja2 3.1.2
joblib 1.1.0
MarkupSafe 2.1.1
ninja 1.10.2.3
numpy 1.21.6
pandas 1.3.5
pip 22.1.2
pyparsing 3.0.9
python-dateutil 2.8.2
pytz 2022.2.1
PyYAML 6.0
requests 2.28.1
scikit-learn 1.0.2
scipy 1.7.3
setuptools 61.2.0
six 1.16.0
threadpoolctl 3.1.0
torch 1.8.0+cu111
torch-geometric 2.0.4
torch-scatter 2.0.8
torch-sparse 0.6.12
tqdm 4.64.0
typing_extensions 4.3.0
urllib3 1.26.11
wheel 0.37.1

Hi, we are facing the following issue while running Linux command line python script/run.py -c config/transductive/wn18rr.yaml --gpus [0]

Traceback (most recent call last):
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1667, in _run_ninja_build
    subprocess.run(
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/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 216, in <module>
    train_and_validate(cfg, model, train_data, valid_data, filtered_data=filtered_data)
  File "script/run.py", line 57, in train_and_validate
    pred = parallel_model(train_data, batch)
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/models.py", line 139, in forward
    output = self.bellmanford(data, h_index[:, 0], r_index[:, 0])  # (num_nodes, batch_size, feature_dim）
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/models.py", line 104, in bellmanford
    hidden = layer(layer_input, query, boundary, data.edge_index, data.edge_type, size, edge_weight)
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/layers.py", line 65, in forward
    output = self.propagate(input=input, relation=relation, boundary=boundary, edge_index=edge_index,
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/layers.py", line 89, in propagate
    out = self.message_and_aggregate(edge_index, **msg_aggr_kwargs)
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/layers.py", line 157, in message_and_aggregate
    from .rspmm import generalized_rspmm
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/rspmm/__init__.py", line 1, in <module>
    from .rspmm import generalized_rspmm
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/rspmm/rspmm.py", line 204, in <module>
    rspmm = load_extension("rspmm", [os.path.join(path, "rspmm.cpp"), os.path.join(path, "rspmm.cu")])
  File "/export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/rspmm/rspmm.py", line 199, in load_extension
    return cpp_extension.load(name, sources, extra_cflags, extra_cuda_cflags, **kwargs)
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1079, in load
    return _jit_compile(
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1292, in _jit_compile
    _write_ninja_file_and_build_library(
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1404, in _write_ninja_file_and_build_library
    _run_ninja_build(
  File "/export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1683, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'rspmm': [1/2] /cm/shared/apps/cuda11.1/toolkit/11.1.0/bin/nvcc --generate-dependencies-with-compile --dependency-output rspmm.cuda.o.d -ccbin /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/bin/x86_64-conda_cos6-linux-gnu-cc -DTORCH_EXTENSION_NAME=rspmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/TH -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/THC -isystem /cm/shared/apps/cuda11.1/toolkit/11.1.0/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/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 /export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/rspmm/source/rspmm.cu -o rspmm.cuda.o
FAILED: rspmm.cuda.o
/cm/shared/apps/cuda11.1/toolkit/11.1.0/bin/nvcc --generate-dependencies-with-compile --dependency-output rspmm.cuda.o.d -ccbin /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/bin/x86_64-conda_cos6-linux-gnu-cc -DTORCH_EXTENSION_NAME=rspmm -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/torch/csrc/api/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/TH -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/lib/python3.8/site-packages/torch/include/THC -isystem /cm/shared/apps/cuda11.1/toolkit/11.1.0/include -isystem /export/home/s190009/myFile/tools/anaconda3/envs/py38pt180cuda111/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 /export/home/s190009/myFile/repository/python/NBFNet-PyG/nbfnet/rspmm/source/rspmm.cu -o rspmm.cuda.o
x86_64-conda_cos6-linux-gnu-cc: error trying to exec 'cc1plus': execvp: Permission denied
nvcc fatal   : Failed to preprocess host compiler properties.
ninja: build stopped: subcommand failed.

We created the required environment, which includes:

python                     3.8.11
torch                        1.8.0+cu111
torch-scatter            2.0.8
torch-sparse            0.6.12
torch-geometric      2.0.1
ninja                        1.10.2.3
easydict                   1.9
pyyaml                    5.4.1

And our gcc version is:

gcc version 7.3.0 (crosstool-NG 1.23.0.449-a04d0)

Could you please check out this issue, and provide some clues on how to resolve it? Thanks.