| /dports/science/py-openpiv/openpiv-python-0.23.8/openpiv/ |
| H A D | lib.py | 72 kernel_size = int(kernel_size)
81 dist, dist_inv = get_dist(kernel, kernel_size)
82 kernel[dist <= kernel_size] = 1
85 kernel[dist <= kernel_size] = dist_inv[dist <= kernel_size]
167 def get_dist(kernel, kernel_size): argument
176 dist = np.sqrt((ys - kernel_size) ** 2 + (xs - kernel_size) ** 2)
177 dist_inv = np.sqrt(2) * kernel_size - dist
182 (ys - kernel_size) ** 2 +
183 (xs - kernel_size) ** 2 +
184 (zs - kernel_size) ** 2
[all …]
|
| /dports/math/gemmlowp/gemmlowp-dc69acd/meta/ |
| H A D | single_thread_transform.h | 24 template <typename Params, int kernel_size>
31 template <typename P, int kernel_size, int leftovers>
34 typename P::Kernel, kernel_size, in ExecuteDispatch1D()
40 template <typename E, typename P, int kernel_size, int variable_leftovers>
59 template <typename E, typename P, int kernel_size>
60 struct Dispatch1D<E, P, kernel_size, 0> {
64 std::cout << "Dispatch(1): " << kernel_size << ": 0" << std::endl
69 E::template ExecuteDispatch1D<P, kernel_size, 0>(params);
80 template <typename Params, int kernel_size>
82 internal::Dispatch1D<internal::Transform1DExecutor, Params, kernel_size,
[all …]
|
| /dports/games/fs2open/fs2open.github.com-release_21_4_1/lib/libRocket/Source/Core/ |
| H A D | ConvolutionFilter.cpp | 36 kernel_size = 0; in ConvolutionFilter()
53 kernel_size = Math::Max(_kernel_size, 1); in Initialise()
54 kernel_size = kernel_size * 2 + 1; in Initialise()
56 kernel = new float[kernel_size * kernel_size]; in Initialise()
57 memset(kernel, 0, kernel_size * kernel_size * sizeof(float)); in Initialise()
69 index = Math::Min(index, kernel_size - 1); in operator []()
71 return kernel + kernel_size * index; in operator []()
84 for (int kernel_y = 0; kernel_y < kernel_size; ++kernel_y) in Run()
86 int source_y = y - source_offset.y - ((kernel_size - 1) / 2) + kernel_y; in Run()
88 for (int kernel_x = 0; kernel_x < kernel_size; ++kernel_x) in Run()
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/python/mxnet/gluon/contrib/cnn/ |
| H A D | conv_layers.py | 113 if isinstance(kernel_size, numeric_types):
114 kernel_size = (kernel_size,) * 2
116 strides = (strides,) * len(kernel_size)
118 padding = (padding,) * len(kernel_size)
123 offset_channels = 2 * kernel_size[0] * kernel_size[1] * num_deformable_group
139 dshape = [0] * (len(kernel_size) + 2)
306 if isinstance(kernel_size, numeric_types):
307 kernel_size = (kernel_size,) * 2
316 offset_channels = num_deformable_group * 3 * kernel_size[0] * kernel_size[1]
317 self.offset_split_index = num_deformable_group * 2 * kernel_size[0] * kernel_size[1]
[all …]
|
| /dports/misc/py-mxnet/incubator-mxnet-1.9.0/python/mxnet/gluon/contrib/cnn/ |
| H A D | conv_layers.py | 113 if isinstance(kernel_size, numeric_types):
114 kernel_size = (kernel_size,) * 2
116 strides = (strides,) * len(kernel_size)
118 padding = (padding,) * len(kernel_size)
123 offset_channels = 2 * kernel_size[0] * kernel_size[1] * num_deformable_group
139 dshape = [0] * (len(kernel_size) + 2)
306 if isinstance(kernel_size, numeric_types):
307 kernel_size = (kernel_size,) * 2
316 offset_channels = num_deformable_group * 3 * kernel_size[0] * kernel_size[1]
317 self.offset_split_index = num_deformable_group * 2 * kernel_size[0] * kernel_size[1]
[all …]
|
| /dports/devel/boost-docs/boost_1_72_0/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/math/stanmath/math-4.2.0/lib/boost_1.75.0/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/devel/boost-libs/boost_1_72_0/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/devel/boost-python-libs/boost_1_72_0/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/science/py-scipy/scipy-1.7.1/scipy/_lib/boost/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/devel/hyperscan/boost_1_75_0/boost/gil/image_processing/ |
| H A D | filter.hpp | 31 std::size_t kernel_size, in box_filter() argument
46 if (normalize) { kernel_values.resize(kernel_size, 1.0f / float(kernel_size)); } in box_filter()
47 else { kernel_values.resize(kernel_size, 1.0f); } in box_filter()
49 if (anchor == -1) anchor = static_cast<int>(kernel_size / 2); in box_filter()
62 std::size_t kernel_size, in blur() argument
76 std::size_t half_kernel_size = kernel_size / 2; in filter_median_impl()
82 values.reserve(kernel_size * kernel_size); in filter_median_impl()
93 kernel_size, in filter_median_impl()
94 kernel_size in filter_median_impl()
114 std::size_t half_kernel_size = kernel_size / 2; in median_filter()
[all …]
|
| /dports/misc/py-mxnet/incubator-mxnet-1.9.0/tests/python/gpu/ |
| H A D | test_gluon_contrib_gpu.py | 32 DeformableConvolution(10, kernel_size=(3, 3), strides=1, padding=0),
33 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
35 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
37 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
40 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, offset_use_bias=False),
41 DeformableConvolution(12, kernel_size=(3, 2), strides=1, padding=0, use_bias=False),
66 DeformableConvolution(10, kernel_size=(3, 3), strides=1, padding=0),
67 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
69 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
71 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/tests/python/gpu/ |
| H A D | test_gluon_contrib_gpu.py | 32 DeformableConvolution(10, kernel_size=(3, 3), strides=1, padding=0),
33 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
35 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
37 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
40 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, offset_use_bias=False),
41 DeformableConvolution(12, kernel_size=(3, 2), strides=1, padding=0, use_bias=False),
66 DeformableConvolution(10, kernel_size=(3, 3), strides=1, padding=0),
67 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
69 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
71 DeformableConvolution(10, kernel_size=(3, 2), strides=1, padding=0, activation='relu',
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/python/mxnet/gluon/nn/ |
| H A D | conv_layers.py | 119 dshape = [-1]*(len(kernel_size) + 2)
121 dshape = [0]*(len(kernel_size) + 2)
247 if isinstance(kernel_size, numeric_types):
248 kernel_size = (kernel_size,)
331 if isinstance(kernel_size, numeric_types):
332 kernel_size = (kernel_size,)*2
416 if isinstance(kernel_size, numeric_types):
417 kernel_size = (kernel_size,)*3
502 kernel_size = (kernel_size,)
596 kernel_size = (kernel_size,)*2
[all …]
|
| /dports/misc/py-mxnet/incubator-mxnet-1.9.0/python/mxnet/gluon/nn/ |
| H A D | conv_layers.py | 119 dshape = [-1]*(len(kernel_size) + 2)
121 dshape = [0]*(len(kernel_size) + 2)
247 if isinstance(kernel_size, numeric_types):
248 kernel_size = (kernel_size,)
331 if isinstance(kernel_size, numeric_types):
332 kernel_size = (kernel_size,)*2
416 if isinstance(kernel_size, numeric_types):
417 kernel_size = (kernel_size,)*3
502 kernel_size = (kernel_size,)
596 kernel_size = (kernel_size,)*2
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/src/relay/op/nn/ |
| H A D | convolution.h | 92 if (param->kernel_size.defined()) { in Conv1DRel()
181 param->kernel_size[1]}}; in Conv2DRel()
201 if (param->kernel_size.defined()) { in Conv2DRel()
289 param->kernel_size[1], param->kernel_size[2]}}; in Conv3DRel()
292 param->kernel_size[1], param->kernel_size[2]}}; in Conv3DRel()
311 if (param->kernel_size.defined()) { in Conv3DRel()
865 param->kernel_size[0], param->kernel_size[1], param->kernel_size[2]}); in Conv3DTransposeRel()
962 param->kernel_size[0], param->kernel_size[1]}); in Conv2DTransposeRel()
1031 param->kernel_size[0], param->kernel_size[1]}); in DeformableConv2DRel()
1033 ksize_y = param->kernel_size[0]; in DeformableConv2DRel()
[all …]
|
| /dports/audio/osd-lyrics/osdlyrics-0.4.3/src/ |
| H A D | ol_gussian_blur.c | 96 int kernel_size = ceil (sigma * 6); in _calc_kernel() local
97 if (kernel_size % 2 == 0) in _calc_kernel()
98 kernel_size++; in _calc_kernel()
99 int orig = kernel_size / 2; in _calc_kernel()
100 if (size) *size = kernel_size; in _calc_kernel()
103 int *kernel = g_new (int, kernel_size); in _calc_kernel()
107 for (i = 0; i < kernel_size; i++) in _calc_kernel()
113 for (i = 0; i < kernel_size; i++) in _calc_kernel()
125 ol_assert (kernel_size > 0 && kernel_size % 2 == 1); in _apply_kernel()
136 int kernel_orig = kernel_size / 2; in _apply_kernel()
[all …]
|
| /dports/science/py-chainer/chainer-7.8.0/chainerx_cc/chainerx/routines/ |
| H A D | pooling.cc | 28 void CheckPoolInputs(const Array& x, const Dims& kernel_size, const Dims& stride, const Dims& pad) { in CheckPoolInputs() argument
30 if (static_cast<int8_t>(kernel_size.size()) != ndim) { in CheckPoolInputs()
42 if (std::any_of(kernel_size.begin(), kernel_size.end(), [](int64_t ks) { return ks <= 0; })) { in CheckPoolInputs()
53 CheckPoolInputs(x, kernel_size, stride, pad); in MaxPool()
75 gout.AsGradStopped(), kernel_size, stride, pad, state, true, absl::nullopt); in MaxPool()
108 Dims kernel_size; in MaxPool() member
118 bt1.Define(MaxPoolBwd{kernel_size, stride, pad, cover_all, std::move(state)}); in MaxPool()
129 CheckPoolInputs(x, kernel_size, stride, pad); in AveragePool()
136 x.AsGradStopped(), kernel_size, stride, pad, pad_mode, true, absl::nullopt); in AveragePool()
156 bt2.Define([kernel_size, stride, pad, pad_mode](BackwardContext& bctx2) { in AveragePool()
[all …]
|
| /dports/misc/py-gluoncv/gluon-cv-0.9.0/gluoncv/model_zoo/siamrpn/ |
| H A D | siam_rpn.py | 42 def __init__(self, hidden, out_channels, bz=1, is_train=False, kernel_size=3, ctx=cpu()): argument
48 self.conv_kernel.add(nn.Conv2D(hidden, kernel_size=kernel_size, use_bias=False),
51 self.conv_search.add(nn.Conv2D(hidden, kernel_size=kernel_size, use_bias=False),
54 self.head.add(nn.Conv2D(hidden, kernel_size=1, use_bias=False),
57 nn.Conv2D(out_channels, kernel_size=1))
59 self.kernel_size = [bz, 256, 4, 4]
63 self.kernel_size = [1, 256, 4, 4]
75 batch = self.kernel_size[0]
76 channel = self.kernel_size[1]
78 kernel = kernel.reshape((batch*channel, 1, self.kernel_size[2], self.kernel_size[3]))
[all …]
|
| /dports/graphics/libjxl/libjxl-0.6.1/tools/upscaling_coefficients/ |
| H A D | generate_upscaling_coefficients.py | 115 patch_size = 2 * kernel_size + 1
127 lower = middle - kernel_size // 2
128 upper = middle + kernel_size // 2 + 1
129 assert len(range(lower, upper)) == kernel_size
148 def indices_matrix(upscaling_factor, kernel_size=5): argument
154 for i in range((kernel_size * upscaling_factor) // 2):
175 for j in range(kernel_size):
178 for b in range(kernel_size):
188 def weights_arrays(upscaling_factor, kernel_size=5): argument
196 for b in range(kernel_size)) + "}"
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/3rdparty/tvm/python/tvm/relay/op/nn/ |
| H A D | nn.py | 35 kernel_size=None, argument
110 kernel_size = (kernel_size,)
124 kernel_size,
140 kernel_size=None, argument
215 kernel_size = (kernel_size, kernel_size)
231 kernel_size,
247 kernel_size=None, argument
322 kernel_size = (kernel_size, kernel_size, kernel_size)
336 kernel_size,
421 kernel_size,
[all …]
|
| /dports/misc/mxnet/incubator-mxnet-1.9.0/example/gluon/sn_gan/ |
| H A D | model.py | 34 def __init__(self, num_filter, kernel_size, argument
41 self.kernel_size = kernel_size
51 num_filter, in_channels, kernel_size, kernel_size))
81 kernel=(self.kernel_size, self.kernel_size),
95 channels=512, kernel_size=4, strides=1, padding=0, use_bias=False))
100 channels=256, kernel_size=4, strides=2, padding=1, use_bias=False))
105 channels=128, kernel_size=4, strides=2, padding=1, use_bias=False))
110 channels=64, kernel_size=4, strides=2, padding=1, use_bias=False))
125 … d_net.add(SNConv2D(num_filter=64, kernel_size=4, strides=2, padding=1, in_channels=3, ctx=ctx))
128 … d_net.add(SNConv2D(num_filter=128, kernel_size=4, strides=2, padding=1, in_channels=64, ctx=ctx))
[all …]
|
| /dports/misc/py-mxnet/incubator-mxnet-1.9.0/example/gluon/sn_gan/ |
| H A D | model.py | 34 def __init__(self, num_filter, kernel_size, argument
41 self.kernel_size = kernel_size
51 num_filter, in_channels, kernel_size, kernel_size))
81 kernel=(self.kernel_size, self.kernel_size),
95 channels=512, kernel_size=4, strides=1, padding=0, use_bias=False))
100 channels=256, kernel_size=4, strides=2, padding=1, use_bias=False))
105 channels=128, kernel_size=4, strides=2, padding=1, use_bias=False))
110 channels=64, kernel_size=4, strides=2, padding=1, use_bias=False))
125 … d_net.add(SNConv2D(num_filter=64, kernel_size=4, strides=2, padding=1, in_channels=3, ctx=ctx))
128 … d_net.add(SNConv2D(num_filter=128, kernel_size=4, strides=2, padding=1, in_channels=64, ctx=ctx))
[all …]
|
| /dports/graphics/py-scikit-image/scikit-image-0.19.0/skimage/exposure/ |
| H A D | _adapthist.py | 29 def equalize_adapthist(image, kernel_size=None, argument
86 if kernel_size is None:
88 elif isinstance(kernel_size, numbers.Number):
89 kernel_size = (kernel_size,) * image.ndim
90 elif len(kernel_size) != image.ndim:
93 kernel_size = [int(k) for k in kernel_size]
95 image = _clahe(image, kernel_size, clip_limit, nbins)
100 def _clahe(image, kernel_size, clip_limit, nbins): argument
131 pad_start_per_dim = [k // 2 for k in kernel_size]
154 for k, n in zip(kernel_size, ns_hist)]
[all …]
|
| /dports/misc/py-tvm/incubator-tvm-0.6.1/src/relay/op/nn/ |
| H A D | convolution.cc | 74 attrs->kernel_size = std::move(kernel_size); in MakeConv2D()
157 param->kernel_size[0], in Conv2DTransposeRel()
223 attrs->kernel_size = std::move(kernel_size); in MakeConv2DTranspose()
378 attrs->kernel_size = std::move(kernel_size); in MakeConv2DWinograd()
490 attrs->kernel_size = std::move(kernel_size); in MakeConv2DWinogradNNPACK()
603 attrs->kernel_size = std::move(kernel_size); in MakeConv2DNCHWcInt8()
652 attrs->kernel_size = std::move(kernel_size); in MakeConv2DNCHWc()
702 attrs->kernel_size = std::move(kernel_size); in MakeDepthwiseConv2DNCHWc()
752 param->kernel_size[0], in DeformableConv2DRel()
753 param->kernel_size[1]}); in DeformableConv2DRel()
[all …]
|