MCPcopy Create free account
hub / github.com/CVCUDA/CV-CUDA / ExportOpConv2D

Function ExportOpConv2D

python/mod_cvcuda/operators/OpConv2D.cpp:74–130  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

72} // namespace
73
74void ExportOpConv2D(py::module &m)
75{
76 using namespace pybind11::literals;
77
78 m.def("conv2d", &Conv2DVarShape, "src"_a, "kernel"_a, "kernel_anchor"_a,
79 "border"_a = NVCVBorderType::NVCV_BORDER_CONSTANT, py::kw_only(), "stream"_a = nullptr, R"pbdoc(
80
81 Executes the Convolve 2D operation on the given cuda stream.
82
83 See also:
84 Refer to the CV-CUDA C API reference for the Convolve 2D operator
85 for more details and usage examples.
86
87 Args:
88 src (cvcuda.ImageBatchVarShape): Input image batch containing one or more images.
89 kernel (cvcuda.Tensor): Convolution kernels (one for each batch image) to be used. Each image width and height
90 correspond to the kernel width and height. (must be float)
91 kernel_anchor (cvcuda.Tensor): 1D Tensor with the anchor of each kernel (one for each batch image). The anchor (x, y)
92 indicates the relative position of a filtered point within the kernel.
93 (-1, -1) means that the anchor is at the kernel center.
94 stream (cvcuda.Stream, optional): CUDA Stream on which to perform the operation.
95
96 Returns:
97 cvcuda.ImageBatchVarShape: The output image batch.
98
99 Caution:
100 Restrictions to several arguments may apply. Check the C
101 API references of the CV-CUDA operator.
102 )pbdoc");
103
104 m.def("conv2d_into", &Conv2DVarShapeInto, "dst"_a, "src"_a, "kernel"_a, "kernel_anchor"_a,
105 "border"_a = NVCVBorderType::NVCV_BORDER_CONSTANT, py::kw_only(), "stream"_a = nullptr, R"pbdoc(
106
107 Executes the Convolve 2D operation on the given cuda stream.
108
109 See also:
110 Refer to the CV-CUDA C API reference for the Convolve 2D operator
111 for more details and usage examples.
112
113 Args:
114 dst (cvcuda.ImageBatchVarShape): Output image batch to store the result of the operation.
115 src (cvcuda.ImageBatchVarShape): Input image batch containing one or more images.
116 kernel (cvcuda.Tensor): Convolution kernels (one for each batch image) to be used. Each image width and height
117 correspond to the kernel width and height. (must be float)
118 kernel_anchor (cvcuda.Tensor): 1D Tensor with the anchor of each kernel (one for each batch image). The anchor (x, y)
119 indicates the relative position of a filtered point within the kernel.
120 (-1, -1) means that the anchor is at the kernel center.
121 stream (cvcuda.Stream, optional): CUDA Stream on which to perform the operation.
122
123 Returns:
124 None
125
126 Caution:
127 Restrictions to several arguments may apply. Check the C
128 API references of the CV-CUDA operator.
129 )pbdoc");
130}
131

Callers 1

PYBIND11_MODULEFunction · 0.85

Calls

no outgoing calls

Tested by

no test coverage detected