| /external/autotest/client/common_lib/cros/ |
| perf_stat_lib.py | 73 variance = sum([(elem - mean) ** 2 for elem in num_list]) / (n -1)
74 return round(sqrt(variance), 2
|
| /external/tensorflow/tensorflow/core/api_def/base_api/ |
| api_def_BatchNormWithGlobalNormalizationGrad.pbtxt | 20 A 1D variance Tensor with size matching the last dimension of t.
54 1D backprop tensor for variance.
|
| api_def_BatchNormWithGlobalNormalization.pbtxt | 20 A 1D variance Tensor with size matching the last dimension of t.
|
| /external/libvpx/libvpx/vp9/encoder/ |
| vp9_blockiness.c | 23 static int variance(int sum, int sum_squared, int size) { function
45 // by dividing the blockiness by the variance of the pixels on either side
70 var_0 = variance(sum_0, sum_sq_0, size);
71 var_1 = variance(sum_1, sum_sq_1, size);
102 var_0 = variance(sum_0, sum_sq_0, size);
103 var_1 = variance(sum_1, sum_sq_1, size);
|
| vp9_noise_estimate.c | 173 // the variance to update estimate of noise in the source.
220 // Compute variance.
221 unsigned int variance = cpi->fn_ptr[bsize].vf( local
224 // average term (sse - variance = N * avg^{2}, N = 16X16) of the
227 if ((sse - variance) < thresh_sum_diff) {
231 // Avoid blocks with high brightness and high spatial variance.
234 avg_est += low_res ? variance >> 4
235 : variance / ((spatial_variance >> 9) + 1);
|
| /packages/apps/Dialer/java/com/android/incallui/answer/impl/classifier/ |
| AnglesClassifier.java | 26 * A classifier which calculates the variance of differences between successive angles in a stroke.
30 * calculated angle. Then it calculates the variance of the differences from a stroke. To the
36 * biggest angle is. It calculates the angle variance of the two parts and sums them up. The reason
39 * The final result is the minimum of angle variance of the whole stroke and the sum of angle
147 // the angle variance so far and start to count the values for the angle
148 // variance of the second part.
|
| /external/python/cpython3/Lib/ |
| statistics.py | 5 averages, variance, and standard deviation.
52 pvariance Population variance of data.
53 variance Sample variance of data.
80 'pstdev', 'pvariance', 'stdev', 'variance',
515 # See http://mathworld.wolfram.com/Variance.html
520 # variance", as that is only suitable for hand calculations with a small
546 def variance(data, xbar=None): function
547 """Return the sample variance of data.
554 calculate the variance from the entire population, see ``pvariance``
[all...] |
| /external/tensorflow/tensorflow/core/kernels/ |
| fused_batch_norm_op.cc | 100 Eigen::Tensor<U, 1, Eigen::RowMajor> variance(depth);
113 variance.device(d) = x_centered.square().sum(reduce_dims) * rest_size_inv;
114 batch_var.device(d) = variance * rest_size_adjust;
115 saved_var.device(d) = variance;
117 variance.device(d) = estimated_variance;
120 auto scaling_factor = ((variance + epsilon).rsqrt() * scale)
148 typename TTypes<U>::ConstVec variance(variance_input.vec<U>());
155 // x_backprop = scale * rsqrt(variance + epsilon) *
157 // mean(y_backprop * (x - mean(x))) / (variance + epsilon)]
159 // (x - mean(x)) * rsqrt(variance + epsilon)
[all...] |
| /frameworks/base/tests/JankBench/scripts/external/ |
| statistics.py | 22 averages, variance, and standard deviation.
68 pvariance Population variance of data.
69 variance Sample variance of data.
96 'pstdev', 'pvariance', 'stdev', 'variance',
482 # See http://mathworld.wolfram.com/Variance.html
487 # variance", as that is only suitable for hand calculations with a small
513 def variance(data, xbar=None): function
514 """Return the sample variance of data.
521 calculate the variance from the entire population, see ``pvariance``
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/regression/ |
| AbstractMultipleLinearRegression.java | 25 import org.apache.commons.math.stat.descriptive.moment.Variance;
289 * Estimates the variance of the error.
291 * @return estimate of the error variance
317 * Calculates the beta variance of multiple linear regression in matrix
320 * @return beta variance
326 * Calculates the variance of the y values.
328 * @return Y variance
331 return new Variance().evaluate(Y.getData());
335 * <p>Calculates the variance of the error term.</p>
342 * @return error variance estimat
[all...] |
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/ |
| StatUtils.java | 25 import org.apache.commons.math.stat.descriptive.moment.Variance;
63 /** variance */
64 private static final Variance VARIANCE = new Variance();
301 * Returns the variance of the entries in the input array, or
304 * See {@link org.apache.commons.math.stat.descriptive.moment.Variance} for
312 * @return the variance of the values or Double.NaN if the array is empty
315 public static double variance(final double[] values) { method in class:StatUtils
316 return VARIANCE.evaluate(values)
339 public static double variance(final double[] values, final int begin, method in class:StatUtils
370 public static double variance(final double[] values, final double mean, method in class:StatUtils
397 public static double variance(final double[] values, final double mean) { method in class:StatUtils
[all...] |
| /external/tensorflow/tensorflow/contrib/timeseries/examples/ |
| predict.py | 61 # variance. SQUARED_LOSS overestimates variance when there are trends in
89 variance = np.squeeze(np.concatenate(
92 upper_limit = mean + np.sqrt(variance)
93 lower_limit = mean - np.sqrt(variance)
|
| /external/webrtc/webrtc/common_audio/ |
| audio_converter_unittest.cc | 58 float variance = 0; local
64 variance += ref.channels()[i][j] * ref.channels()[i][j];
71 variance /= length;
73 variance -= mean * mean;
76 snr = 10 * std::log10(variance / mse);
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/inference/ |
| TTestImpl.java | 191 return t(StatUtils.mean(observed), mu, StatUtils.variance(observed),
235 * and <strong><code>var</code></strong> is the pooled variance estimate:
239 * with <strong><code>var1<code></strong> the variance of the first sample and
240 * <strong><code>var2</code></strong> the variance of the second sample.
256 StatUtils.variance(sample1), StatUtils.variance(sample2),
276 * <strong><code> var1</code></strong> is the variance of the first sample;
277 * <strong><code> var2</code></strong> is the variance of the second sample;
293 StatUtils.variance(sample1), StatUtils.variance(sample2)
[all...] |
| /external/libchrome/base/trace_event/ |
| memory_dump_scheduler.cc | 249 uint64_t variance = 0; local
251 variance += (polling_state_->last_memory_totals_kb[i] - mean) *
254 variance = variance / PollingTriggerState::kMaxNumMemorySamples;
261 bool is_stddev_low = variance < mean / 500 * mean / 500;
268 (3.69 * 3.69 * variance);
|
| /external/tensorflow/tensorflow/compiler/xla/service/gpu/ |
| cudnn_batchnorm_rewriter.cc | 121 // {output, mean, rsqrt(variance + epsilon)},
122 // but the batchnorm HLO returns {output, mean, variance}. Fix it up.
131 HloInstruction* variance = local
142 variance,
173 // The cudnn libcall expects its input to be rsqrt(variance + epsilon), but
174 // the batchnorm HLO takes plain variance as input. Fix it up.
|
| /frameworks/base/packages/SystemUI/src/com/android/systemui/classifier/ |
| AnglesClassifier.java | 29 * A classifier which calculates the variance of differences between successive angles in a stroke.
33 * previously calculated angle. Then it calculates the variance of the differences from a stroke.
39 * angle is. It calculates the angle variance of the two parts and sums them up. The reason the
42 * final result is the minimum of angle variance of the whole stroke and the sum of angle variances
152 // the angle variance so far and start to count the values for the angle
153 // variance of the second part.
|
| /external/tensorflow/tensorflow/python/ops/ |
| nn_impl.py | 571 """Calculate the sufficient statistics for the mean and variance of `x`.
579 axes: Array of ints. Axes along which to compute mean and variance.
621 """Calculate the mean and variance of based on the sufficient statistics.
627 variance_ss: A `Tensor` containing the variance sufficient statistics: the
628 (possibly shifted) squared sum of the data to compute the variance over.
634 Two `Tensor` objects: `mean` and `variance`.
644 variance = math_ops.subtract(
647 name="variance")
648 return (mean, variance)
658 """Calculate the mean and variance of `x`
[all...] |
| /cts/apps/CameraITS/tests/scene0/ |
| test_jitter.py | 31 MAX_VAR_FRAME_DELTA = 0.01 # variance of frame deltas
51 print "Variance:", var
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/distribution/ |
| WeibullDistributionImpl.java | 61 /** Cached numerical variance */
64 /** Whether or not the numerical variance has been calculated */
320 * Calculates the variance.
322 * The variance is
326 * @return the variance
355 * Returns the variance of the distribution.
357 * @return the variance (possibly Double.POSITIVE_INFINITY as
372 * Invalidates the cached mean and variance.
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/stat/descriptive/ |
| StatisticalSummary.java | 33 * Returns the variance of the available values.
34 * @return The variance, Double.NaN if no values have been added
|
| /external/opencv/cvaux/src/ |
| cvbgfg_gaussmix.cpp | 79 icvMatchTest(...) assumes what all color channels component exhibit the same variance
120 //Rw is the learning rate for weight and Rg is leaning rate for mean and variance
126 //The list is maintained in sorted order using w/sqrt(variance) as a key
132 //v[n+1] = v[n] + Rg*((x[n+1] - u[n])*(x[n+1] - u[n])) - v[n]) variance
206 bg_model->g_point[n].g_values[0].variance[m] = var_init;
214 bg_model->g_point[n].g_values[k].variance[m] = var_init;
390 var_threshold += g_point->g_values[k].variance[m];
420 sum_d2 += (d*d) / (g_point->g_values[k].variance[m] * g_point->g_values[k].variance[m]);
452 g_point->g_values[k].variance[m] = g_point->g_values[k].variance[m]
[all...] |
| /external/tensorflow/tensorflow/python/layers/ |
| normalization.py | 64 epsilon: Small float added to variance to avoid dividing by zero.
73 moving_variance_initializer: Initializer for the moving variance.
301 # Disable variable partitioning when creating the moving mean and variance
395 variance=self.moving_variance,
400 output, mean, variance = utils.smart_cond(
404 # Note that the variance computed by fused batch norm is
407 array_ops.size(inputs) / array_ops.size(variance), variance.dtype)
408 factor = (sample_size - math_ops.cast(1.0, variance.dtype)) / sample_size
409 variance *= facto
[all...] |
| /external/libmpeg2/common/ |
| ideint_structs.h | 57 /** Adaptive variance used in spatio temporal filtering */
|
| /external/tensorflow/tensorflow/contrib/layers/python/layers/ |
| normalization.py | 69 epsilon: Small float added to variance to avoid dividing by zero.
73 moving variance.
153 mean, variance = nn.moments(inputs, moments_axes, keep_dims=True)
157 inputs, mean, variance, beta, gamma, epsilon, name='instancenorm')
|
