| /external/pytorch/tools/autograd/ |
| D | deprecated.yaml | 13 - name: addbmm(Scalar beta, Tensor self, Scalar alpha, Tensor batch1, Tensor batch2) -> Tensor
14 aten: addbmm(self, batch1, batch2, beta, alpha)
16 - name: addbmm_(Scalar beta, Tensor(a!) self, Scalar alpha, Tensor batch1, Tensor batch2) -> Tensor…
17 aten: addbmm_(self, batch1, batch2, beta, alpha)
19 - name: addbmm(Scalar beta, Tensor self, Scalar alpha, Tensor batch1, Tensor batch2, *, Tensor(a!) …
20 aten: addbmm_out(out, self, batch1, batch2, beta, alpha)
22 - name: addbmm(Scalar beta, Tensor self, Tensor batch1, Tensor batch2) -> Tensor
23 aten: addbmm(self, batch1, batch2, beta, 1)
25 - name: addbmm_(Scalar beta, Tensor(a!) self, Tensor batch1, Tensor batch2) -> Tensor(a!)
26 aten: addbmm_(self, batch1, batch2, beta, 1)
[all …]
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math3/distribution/ |
| D | BetaDistribution.java | 23 import org.apache.commons.math3.special.Beta;
29 * Implements the Beta distribution.
31 * @see <a href="http://en.wikipedia.org/wiki/Beta_distribution">Beta distribution</a>
49 private final double beta; field in BetaDistribution
52 * Normalizing factor used in density computations. updated whenever alpha or beta are changed.
69 * @param beta Second shape parameter (must be positive).
71 public BetaDistribution(double alpha, double beta) { in BetaDistribution() argument
72 this(alpha, beta, DEFAULT_INVERSE_ABSOLUTE_ACCURACY); in BetaDistribution()
85 * @param beta Second shape parameter (must be positive).
90 public BetaDistribution(double alpha, double beta, double inverseCumAccuracy) { in BetaDistribution() argument
[all …]
|
| D | LaplaceDistribution.java | 42 private final double beta; field in LaplaceDistribution
54 * @param beta scale parameter (must be positive)
55 * @throws NotStrictlyPositiveException if {@code beta <= 0}
57 public LaplaceDistribution(double mu, double beta) { in LaplaceDistribution() argument
58 this(new Well19937c(), mu, beta); in LaplaceDistribution()
66 * @param beta scale parameter (must be positive)
67 * @throws NotStrictlyPositiveException if {@code beta <= 0}
69 public LaplaceDistribution(RandomGenerator rng, double mu, double beta) { in LaplaceDistribution() argument
72 if (beta <= 0.0) { in LaplaceDistribution()
73 throw new NotStrictlyPositiveException(LocalizedFormats.NOT_POSITIVE_SCALE, beta); in LaplaceDistribution()
[all …]
|
| D | GumbelDistribution.java | 51 private final double beta; field in GumbelDistribution
63 * @param beta scale parameter (must be positive)
64 * @throws NotStrictlyPositiveException if {@code beta <= 0}
66 public GumbelDistribution(double mu, double beta) { in GumbelDistribution() argument
67 this(new Well19937c(), mu, beta); in GumbelDistribution()
75 * @param beta scale parameter (must be positive)
76 * @throws NotStrictlyPositiveException if {@code beta <= 0}
78 public GumbelDistribution(RandomGenerator rng, double mu, double beta) { in GumbelDistribution() argument
81 if (beta <= 0) { in GumbelDistribution()
82 throw new NotStrictlyPositiveException(LocalizedFormats.SCALE, beta); in GumbelDistribution()
[all …]
|
| /external/libaom/tools/ |
| D | gen_constrained_tokenset.py | 16 cdf(x) = 0.5 + 0.5 * sgn(x) * [1 - {alpha/(alpha + |x|)} ^ beta]
18 For a given beta and a given probability of the 1-node, the alpha
19 is first solved, and then the {alpha, beta} pair is used to generate
30 def cdf_spareto(x, xm, beta): argument
31 p = 1 - (xm / (np.abs(x) + xm))**beta
36 def get_spareto(p, beta): argument
40 return ((cdf(1.5, x, beta) - cdf(0.5, x, beta)) /
41 (1 - cdf(0.5, x, beta)) - p)**2
45 parray[0] = 2 * (cdf(0.5, alpha, beta) - 0.5)
46 parray[1] = (2 * (cdf(1.5, alpha, beta) - cdf(0.5, alpha, beta)))
[all …]
|
| /external/cblas/testing/ |
| D | c_d3chke.c | 32 ALPHA=0.0, BETA=0.0; in F77_d3chke() local
51 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
55 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
59 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
63 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
67 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
71 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
75 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
79 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
83 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_d3chke()
[all …]
|
| D | c_s3chke.c | 32 ALPHA=0.0, BETA=0.0; in F77_s3chke() local
50 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
54 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
58 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
62 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
66 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
70 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
74 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
78 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
82 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_s3chke()
[all …]
|
| D | c_z3chke.c | 33 BETA[2] = {0.0,0.0}, in F77_z3chke() local
53 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
57 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
61 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
65 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
69 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
73 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
77 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
81 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
85 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_z3chke()
[all …]
|
| D | c_c3chke.c | 33 BETA[2] = {0.0,0.0}, in F77_c3chke() local
53 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
57 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
61 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
65 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
69 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
73 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
77 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
81 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
85 ALPHA, A, 1, B, 1, BETA, C, 1 ); in F77_c3chke()
[all …]
|
| /external/pytorch/aten/src/ATen/native/ |
| D | Distributions.h | 376 // Approximate reparameterized gradient of Beta(x,alpha,beta) wrt alpha.
379 C10_DEVICE inline scalar_t _beta_grad_alpha_small(scalar_t x, scalar_t alpha, scalar_t beta) { in _beta_grad_alpha_small() argument
381 - digamma_one<scalar_t, accscalar_t>(alpha + beta) - compat_log(x); in _beta_grad_alpha_small()
386 numer *= (casted_i - beta) * x / casted_i; in _beta_grad_alpha_small()
390 const scalar_t result = x * compat_pow(1 - x, -beta) * series; in _beta_grad_alpha_small()
394 // Approximate reparameterized gradient of Beta(x,alpha,beta) wrt beta.
397 C10_DEVICE inline scalar_t _beta_grad_beta_small(scalar_t x, scalar_t alpha, scalar_t beta) { in _beta_grad_beta_small() argument
398 … factor = digamma_one<scalar_t, accscalar_t>(alpha + beta) - digamma_one<scalar_t, accscalar_t>(be… in _beta_grad_beta_small()
403 dbetas = dbetas * (beta - casted_i) + betas; in _beta_grad_beta_small()
404 betas = betas * (beta - casted_i); in _beta_grad_beta_small()
[all …]
|
| /external/tensorflow/tensorflow/python/ops/distributions/ |
| D | beta.py | 15 """The Beta distribution class."""
37 "Beta",
46 @tf_export(v1=["distributions.Beta"])
47 class Beta(distribution.Distribution): class
48 """Beta distribution.
50 The Beta distribution is defined over the `(0, 1)` interval using parameters
51 `concentration1` (aka "alpha") and `concentration0` (aka "beta").
58 pdf(x; alpha, beta) = x**(alpha - 1) (1 - x)**(beta - 1) / Z
59 Z = Gamma(alpha) Gamma(beta) / Gamma(alpha + beta)
65 * `concentration0 = beta`,
[all …]
|
| /external/armnn/src/backends/backendsCommon/test/layerTests/ |
| D | SoftmaxTestImpl.hpp | 19 float beta);
25 float beta,
32 float beta);
38 float beta,
45 float beta);
51 float beta,
58 float beta);
64 float beta);
70 float beta);
76 float beta);
[all …]
|
| D | SoftmaxTestImpl.cpp | 63 float beta, in SimpleSoftmaxBaseTestImpl() argument
94 data.m_Parameters.m_Beta = beta; in SimpleSoftmaxBaseTestImpl()
124 float beta) in SimpleSoftmaxTestImpl() argument
129 float x0[4] = { exp((0.f - 1.0f) * beta), exp((1.0f - 1.0f) * beta), in SimpleSoftmaxTestImpl()
130 exp((0.0f - 1.0f) * beta), exp((0.0f - 1.0f) * beta) }; in SimpleSoftmaxTestImpl()
132 float x1[4] = { exp((0.5f - 0.5f) * beta), exp((0.0f - 0.5f) * beta), in SimpleSoftmaxTestImpl()
133 exp((0.0f - 0.5f) * beta), exp((0.0f - 0.5f) * beta) }; in SimpleSoftmaxTestImpl()
145 … SimpleSoftmaxBaseTestImpl<ArmnnType, 2>(workloadFactory, memoryManager, tensorHandleFactory, beta, in SimpleSoftmaxTestImpl()
154 float beta, in SimpleSoftmaxTestImpl() argument
201 … SimpleSoftmaxBaseTestImpl<ArmnnType, 2>(workloadFactory, memoryManager, tensorHandleFactory, beta, in SimpleSoftmaxTestImpl()
[all …]
|
| /external/tensorflow/tensorflow/python/kernel_tests/distributions/ |
| D | beta_test.py | 26 from tensorflow.python.ops.distributions import beta as beta_lib
51 dist = beta_lib.Beta(a, b)
60 dist = beta_lib.Beta(a, b)
69 dist = beta_lib.Beta(a, b)
78 dist = beta_lib.Beta(a, b)
85 dist = beta_lib.Beta(a, b)
92 dist = beta_lib.Beta(a, b, validate_args=True)
109 dist = beta_lib.Beta(a, b)
118 dist = beta_lib.Beta(a, b)
128 dist = beta_lib.Beta(a, b)
[all …]
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/distribution/ |
| D | BetaDistributionImpl.java | 23 import org.apache.commons.math.special.Beta;
27 * Implements the Beta distribution.
32 * Beta distribution</a></li>
54 private double beta; field in BetaDistributionImpl
57 * updated whenever alpha or beta are changed.
67 * @param beta second shape parameter (must be positive)
72 public BetaDistributionImpl(double alpha, double beta, double inverseCumAccuracy) { in BetaDistributionImpl() argument
74 this.beta = beta; in BetaDistributionImpl()
82 * @param beta second shape parameter (must be positive)
84 public BetaDistributionImpl(double alpha, double beta) { in BetaDistributionImpl() argument
[all …]
|
| D | GammaDistributionImpl.java | 48 private double beta; field in GammaDistributionImpl
54 * Create a new gamma distribution with the given alpha and beta values.
56 * @param beta the scale parameter.
58 public GammaDistributionImpl(double alpha, double beta) { in GammaDistributionImpl() argument
59 this(alpha, beta, DEFAULT_INVERSE_ABSOLUTE_ACCURACY); in GammaDistributionImpl()
63 * Create a new gamma distribution with the given alpha and beta values.
65 * @param beta the scale parameter.
70 public GammaDistributionImpl(double alpha, double beta, double inverseCumAccuracy) { in GammaDistributionImpl() argument
73 setBetaInternal(beta); in GammaDistributionImpl()
100 ret = Gamma.regularizedGammaP(alpha, x / beta); in cumulativeProbability()
[all …]
|
| /external/XNNPACK/test/ |
| D | f32-velu.cc | 89 TEST(F32_VELU__NEON_RR2_LUT16_P3_X4, beta) { in TEST() argument
91 for (float beta : std::vector<float>({0.3f, 3.0f})) { in TEST() local
95 .beta(beta) in TEST()
172 TEST(F32_VELU__NEON_RR2_LUT16_P3_X8, beta) { in TEST() argument
174 for (float beta : std::vector<float>({0.3f, 3.0f})) { in TEST() local
178 .beta(beta) in TEST()
255 TEST(F32_VELU__NEON_RR2_LUT16_P3_X12, beta) { in TEST() argument
257 for (float beta : std::vector<float>({0.3f, 3.0f})) { in TEST() local
261 .beta(beta) in TEST()
338 TEST(F32_VELU__NEON_RR2_LUT16_P3_X16, beta) { in TEST() argument
[all …]
|
| /external/tensorflow/tensorflow/python/kernel_tests/nn_ops/ |
| D | lrn_op_test.py | 37 beta=0.5): argument
52 np.power(bias + alpha * np.sum(patch * patch), beta))
62 # random depth_radius, bias, alpha, beta. cuDNN requires depth_radius to
68 # cuDNN requires beta >= 0.01.
69 beta = 0.01 + 2.0 * np.random.rand()
76 beta=beta)
84 beta=beta)
86 print("LRN error for bias ", bias, "alpha ", alpha, " beta ", beta, " is ",
121 beta = 0.404427052
149 beta=beta))
[all …]
|
| /external/pytorch/aten/src/ATen/native/cpu/ |
| D | BlasKernel.cpp | 21 const float beta,
33 const float beta,
109 opmath_t beta, in gemm_notrans_() argument
112 // c *= beta in gemm_notrans_()
113 scale_(m, n, beta, c, ldc); in gemm_notrans_()
145 opmath_t beta, in gemm_notrans_() argument
155 if (beta == opmath_t(0)) { in gemm_notrans_()
158 c[j * ldc + i] = beta * c[j * ldc + i] + alpha * dot; in gemm_notrans_()
171 opmath_t beta, in gemm_transa_() argument
173 // c = alpha * (a.T @ b) + beta * c in gemm_transa_()
[all …]
|
| /external/cronet/tot/third_party/abseil-cpp/absl/random/ |
| D | beta_distribution.h | 37 // Generate a floating-point variate conforming to a Beta distribution:
38 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
39 // where the params alpha and beta are both strictly positive real values.
42 // to 0 or 1, due to numerical errors when alpha and beta are very different.
44 // Usage note: One usage is that alpha and beta are counts of number of
46 // approximating a beta distribution with a Gaussian distribution with the same
48 // smaller of alpha and beta when the number of trials are sufficiently large,
49 // to quantify how far a beta distribution is from the normal distribution.
59 explicit param_type(result_type alpha, result_type beta) in param_type() argument
60 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|
| /external/abseil-cpp/absl/random/ |
| D | beta_distribution.h | 35 // Generate a floating-point variate conforming to a Beta distribution:
36 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
37 // where the params alpha and beta are both strictly positive real values.
40 // to 0 or 1, due to numerical errors when alpha and beta are very different.
42 // Usage note: One usage is that alpha and beta are counts of number of
44 // approximating a beta distribution with a Gaussian distribution with the same
46 // smaller of alpha and beta when the number of trials are sufficiently large,
47 // to quantify how far a beta distribution is from the normal distribution.
57 explicit param_type(result_type alpha, result_type beta) in param_type() argument
58 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|
| /external/openscreen/third_party/abseil/src/absl/random/ |
| D | beta_distribution.h | 35 // Generate a floating-point variate conforming to a Beta distribution:
36 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
37 // where the params alpha and beta are both strictly positive real values.
40 // to 0 or 1, due to numerical errors when alpha and beta are very different.
42 // Usage note: One usage is that alpha and beta are counts of number of
44 // approximating a beta distribution with a Gaussian distribution with the same
46 // smaller of alpha and beta when the number of trials are sufficiently large,
47 // to quantify how far a beta distribution is from the normal distribution.
57 explicit param_type(result_type alpha, result_type beta) in param_type() argument
58 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|
| /external/angle/third_party/abseil-cpp/absl/random/ |
| D | beta_distribution.h | 37 // Generate a floating-point variate conforming to a Beta distribution:
38 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
39 // where the params alpha and beta are both strictly positive real values.
42 // to 0 or 1, due to numerical errors when alpha and beta are very different.
44 // Usage note: One usage is that alpha and beta are counts of number of
46 // approximating a beta distribution with a Gaussian distribution with the same
48 // smaller of alpha and beta when the number of trials are sufficiently large,
49 // to quantify how far a beta distribution is from the normal distribution.
59 explicit param_type(result_type alpha, result_type beta) in param_type() argument
60 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|
| /external/cronet/stable/third_party/abseil-cpp/absl/random/ |
| D | beta_distribution.h | 37 // Generate a floating-point variate conforming to a Beta distribution:
38 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
39 // where the params alpha and beta are both strictly positive real values.
42 // to 0 or 1, due to numerical errors when alpha and beta are very different.
44 // Usage note: One usage is that alpha and beta are counts of number of
46 // approximating a beta distribution with a Gaussian distribution with the same
48 // smaller of alpha and beta when the number of trials are sufficiently large,
49 // to quantify how far a beta distribution is from the normal distribution.
59 explicit param_type(result_type alpha, result_type beta) in param_type() argument
60 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|
| /external/rust/android-crates-io/crates/grpcio-sys/grpc/third_party/abseil-cpp/absl/random/ |
| D | beta_distribution.h | 35 // Generate a floating-point variate conforming to a Beta distribution:
36 // pdf(x) \propto x^(alpha-1) * (1-x)^(beta-1),
37 // where the params alpha and beta are both strictly positive real values.
40 // to 0 or 1, due to numerical errors when alpha and beta are very different.
42 // Usage note: One usage is that alpha and beta are counts of number of
44 // approximating a beta distribution with a Gaussian distribution with the same
46 // smaller of alpha and beta when the number of trials are sufficiently large,
47 // to quantify how far a beta distribution is from the normal distribution.
57 explicit param_type(result_type alpha, result_type beta) in param_type() argument
58 : alpha_(alpha), beta_(beta) { in param_type()
[all …]
|