twodlearn.kernels module¶

class twodlearn.kernels.ConcatOnes(**kargs)[source]¶

Concat ones at the right of input matrix

Tdl autoinitialization with arguments:

Attributes:

class ConcatOnesBasis(**kargs)[source]¶

class twodlearn.kernels.GaussianKernel(l_scale=None, f_scale=None, y_scale=None, feature_ndims=1, **kargs)[source]¶

gaussian kernel calculation between matrices X1 and X2

X1 is a matrix, whose rows represent samples. X2 is a matrix, whose rows represent samples.

\[K(i,j) = (f_{scale}^2) \exp(-0.5 (x_1(i)-x_2(j))^T (l^{-2}) I (x_1(i)-x_2(j)) ) + y_{scale}^2 I\]

Tdl autoinitialization with arguments:

class twodlearn.kernels.PairwiseL2(x1, x2, name='PairwiseL2')[source]¶

Tdl autoinitialization with arguments:

Attributes:

property x1[source]¶: \(x_1\) input for the PairwiseL2 operation \((||x_1 - x_2||^2)\)

property x2[source]¶: \(x_2\) input for the PairwiseL2 operation \((||x_1 - x_2||^2)\)

class twodlearn.kernels.QuadraticFeatures(include_inputs=False, name=None, **kargs)[source]¶

Concat ones at the right of input matrix

Tdl autoinitialization with arguments: