['-O3', '--use_fast_math'] + version_dependent_macros
}))
# Check, if ATen/CUDAGenerator.h is found, otherwise use the new ATen/CUDAGeneratorImpl.h, due to breaking change in https://github.com/pytorch/pytorch/pull/36026
generator_flag = []
torch_dir = torch.__path__[0]
if os.path.exists(os.path.join(torch_dir, 'include', 'ATen',
'CUDAGenerator.h')):
generator_flag = ['-DOLD_GENERATOR']
if "--fast_multihead_attn" in sys.argv:
from torch.utils.cpp_extension import CUDAExtension
sys.argv.remove("--fast_multihead_attn")
from torch.utils.cpp_extension import BuildExtension
cmdclass['build_ext'] = BuildExtension.with_options(use_ninja=False)
if torch.utils.cpp_extension.CUDA_HOME is None:
raise RuntimeError(
"--fast_multihead_attn was requested, but nvcc was not found. Are you sure your environment has nvcc available? If you're installing within a container from https://hub.docker.com/r/pytorch/pytorch, only images whose names contain 'devel' will provide nvcc."
)
else:
# Check, if CUDA11 is installed for compute capability 8.0
cc_flag = []
_, bare_metal_major, _ = get_cuda_bare_metal_version(
cpp_extension.CUDA_HOME)
if int(bare_metal_major) >= 11:
cc_flag.append('-gencode')
cc_flag.append('arch=compute_80,code=sm_80')
subprocess.run([
return extensions
install_requires = ['scipy']
setup_requires = ['pytest-runner']
tests_require = ['pytest', 'pytest-cov']
setup(
name='torch_sparse',
version='0.6.7',
author='Matthias Fey',
author_email='*****@*****.**',
url='https://github.com/rusty1s/pytorch_sparse',
description=('PyTorch Extension Library of Optimized Autograd Sparse '
'Matrix Operations'),
keywords=['pytorch', 'sparse', 'sparse-matrices', 'autograd'],
license='MIT',
python_requires='>=3.6',
install_requires=install_requires,
setup_requires=setup_requires,
tests_require=tests_require,
extras_require={'test': tests_require},
ext_modules=get_extensions() if not BUILD_DOCS else [],
cmdclass={
'build_ext':
BuildExtension.with_options(no_python_abi_suffix=True, use_ninja=False)
},
packages=find_packages(),
)
readme = f.read()
setup(
# Metadata
name=package_name,
version=version,
author="PyTorch Core Team",
author_email="*****@*****.**",
url="https://github.com/pytorch/vision",
description=
"image and video datasets and models for torch deep learning",
long_description=readme,
license="BSD",
# Package info
packages=find_packages(exclude=("test", )),
package_data={
package_name: ["*.dll", "*.dylib", "*.so", "*.categories"]
},
zip_safe=False,
install_requires=requirements,
extras_require={
"scipy": ["scipy"],
},
ext_modules=get_extensions(),
cmdclass={
"build_ext":
BuildExtension.with_options(no_python_abi_suffix=True),
"clean": clean,
},
)
if torch.cuda.is_available() and (CUDA_HOME is not None or ROCM_HOME is not None):
extension = CUDAExtension(
'torch_test_cpp_extension.torch_library', [
'torch_library.cu'
],
extra_compile_args={'cxx': CXX_FLAGS,
'nvcc': ['-O2']})
ext_modules.append(extension)
# todo(mkozuki): Figure out the root cause
if (not IS_WINDOWS) and torch.cuda.is_available() and CUDA_HOME is not None:
cublas_extension = CUDAExtension(
name='torch_test_cpp_extension.cublas_extension',
sources=['cublas_extension.cpp']
)
ext_modules.append(cublas_extension)
cusolver_extension = CUDAExtension(
name='torch_test_cpp_extension.cusolver_extension',
sources=['cusolver_extension.cpp']
)
ext_modules.append(cusolver_extension)
setup(
name='torch_test_cpp_extension',
packages=['torch_test_cpp_extension'],
ext_modules=ext_modules,
include_dirs='self_compiler_include_dirs_test',
cmdclass={'build_ext': BuildExtension.with_options(use_ninja=USE_NINJA)})
from setuptools import setup, find_packages
from torch.utils.cpp_extension import CUDAExtension, BuildExtension
import os
import glob
libname = "torch_batch_svd"
ext_src = glob.glob(os.path.join(libname, 'csrc/*.cpp'))
print(ext_src)
setup(name=libname,
packages=find_packages(exclude=('tests', 'build', 'csrc', 'include', 'torch_batch_svd.egg-info')),
ext_modules=[CUDAExtension(
libname + '._c',
sources=ext_src,
libraries=["cusolver", "cublas"],
extra_compile_args={'cxx': ['-O2', '-I{}'.format('{}/include'.format(libname))],
'nvcc': ['-O2']}
)],
cmdclass={'build_ext': BuildExtension.with_options(use_ninja=False)}
)
headers += ['src/roi_align_cuda.h']
defines += [('WITH_CUDA', None)]
with_cuda = True
this_file = os.path.dirname(os.path.realpath(__file__))
print(this_file)
extra_objects = ['src/roi_align_kernel.cu.o']
extra_objects = [os.path.join(this_file, fname) for fname in extra_objects]
# ffi = create_extension(
# '_ext.roi_align',
# headers=headers,
# sources=sources,
# define_macros=defines,
# relative_to=__file__,
# with_cuda=with_cuda,
# extra_objects=extra_objects
# )
# fix upper torch 1.0.1
ffi = BuildExtension('_ext.roi_align',
headers=headers,
sources=sources,
define_macros=defines,
relative_to=__file__,
with_cuda=with_cuda,
extra_objects=extra_objects)
if __name__ == '__main__':
ffi.build()
setup(
name='odtk',
version='0.2.5',
description='Fast and accurate single shot object detector',
author = 'NVIDIA Corporation',
packages=['retinanet', 'retinanet.backbones'],
ext_modules=[CUDAExtension('retinanet._C',
['csrc/extensions.cpp', 'csrc/engine.cpp', 'csrc/cuda/decode.cu', 'csrc/cuda/decode_rotate.cu', 'csrc/cuda/nms.cu', 'csrc/cuda/nms_iou.cu'],
extra_compile_args={
'cxx': ['-std=c++14', '-O2', '-Wall'],
'nvcc': [
'-std=c++14', '--expt-extended-lambda', '--use_fast_math', '-Xcompiler', '-Wall',
'-gencode=arch=compute_60,code=sm_60', '-gencode=arch=compute_61,code=sm_61',
'-gencode=arch=compute_70,code=sm_70', '-gencode=arch=compute_72,code=sm_72',
'-gencode=arch=compute_75,code=sm_75', '-gencode=arch=compute_75,code=compute_75'
],
},
libraries=['nvinfer', 'nvinfer_plugin', 'nvonnxparser'])
],
cmdclass={'build_ext': BuildExtension.with_options(no_python_abi_suffix=True)},
install_requires=[
'torch>=1.0.0a0',
'torchvision',
'apex @ git+https://github.com/NVIDIA/apex',
'pycocotools @ git+https://github.com/nvidia/cocoapi.git#subdirectory=PythonAPI',
'pillow',
'requests',
],
entry_points = {'console_scripts': ['odtk=retinanet.main:main']}
)
),
]
# Python interface
setup(
name="MinkowskiEngine",
version=find_version("MinkowskiEngine", "__init__.py"),
install_requires=["torch", "numpy"],
packages=[
"MinkowskiEngine", "MinkowskiEngine.utils", "MinkowskiEngine.modules"
],
package_dir={"MinkowskiEngine": "./MinkowskiEngine"},
ext_modules=ext_modules,
include_dirs=[str(SRC_PATH),
str(SRC_PATH / "3rdparty"), *include_dirs],
cmdclass={"build_ext": BuildExtension.with_options(use_ninja=True)},
author="Christopher Choy",
author_email="*****@*****.**",
description="a convolutional neural network library for sparse tensors",
long_description=read("README.md"),
long_description_content_type="text/markdown",
url="https://github.com/NVIDIA/MinkowskiEngine",
keywords=[
"pytorch",
"Minkowski Engine",
"Sparse Tensor",
"Convolutional Neural Networks",
"3D Vision",
"Deep Learning",
],
zip_safe=False,
# It's an old-style class in Python 2.7...
distutils.command.clean.clean.run(self)
setuptools.setup(
name=package_name,
version=version,
author="Christian Puhrsch",
author_email="*****@*****.**",
description="NestedTensors for PyTorch",
long_description=readme,
long_description_content_type="text/markdown",
url="https://github.com/pytorch/nestedtensor",
packages=setuptools.find_packages(),
classifiers=[
"Programming Language :: Python :: 3",
"Operating System :: OS Independent",
],
zip_safe=True,
cmdclass={
"clean":
clean,
"build_ext":
BuildExtension.with_options(no_python_abi_suffix=True,
use_ninja=os.environ.get(
"USE_NINJA", False)),
},
install_requires=requirements,
ext_modules=get_extensions(),
)
from torch.utils.cpp_extension import BuildExtension
sources = ['src/nms.c']
headers = ['src/nms.h']
defines = []
with_cuda = False
if torch.cuda.is_available():
print('Including CUDA code.')
sources += ['src/nms_cuda.c']
headers += ['src/nms_cuda.h']
defines += [('WITH_CUDA', None)]
with_cuda = True
this_file = os.path.dirname(os.path.realpath(__file__))
print(this_file)
extra_objects = ['src/cuda/nms_kernel.cu.o']
extra_objects = [os.path.join(this_file, fname) for fname in extra_objects]
ffi = BuildExtension('_ext.nms',
headers=headers,
sources=sources,
define_macros=defines,
relative_to=__file__,
with_cuda=with_cuda,
extra_objects=extra_objects,
extra_compile_args=['-std=c99'])
if __name__ == '__main__':
ffi.build()
class FBGEMM_GPU_BuildExtension(BuildExtension.with_options(no_python_abi_suffix=True)):
def build_extension(self, ext):
generate_jinja_files()
super().build_extension(ext)
extra_objects += ['src/cuda/dcn_v2_psroi_pooling_cuda_double.cu.o']
with_cuda = True
else:
raise ValueError('CUDA is not available')
extra_compile_args = ['-fopenmp', '-std=c99']
this_file = os.path.dirname(os.path.realpath(__file__))
print(this_file)
sources = [os.path.join(this_file, fname) for fname in sources]
headers = [os.path.join(this_file, fname) for fname in headers]
extra_objects = [os.path.join(this_file, fname) for fname in extra_objects]
ffi = BuildExtension('_ext.dcn_v2_double',
headers=headers,
sources=sources,
define_macros=defines,
relative_to=__file__,
with_cuda=with_cuda,
extra_objects=extra_objects,
extra_compile_args=extra_compile_args)
# ffi = create_extension(
# '_ext.dcn_v2_double',
# headers=headers,
# sources=sources,
# define_macros=defines,
# relative_to=__file__,
# with_cuda=with_cuda,
# extra_objects=extra_objects,
# extra_compile_args=extra_compile_args
# )
if __name__ == '__main__':
