netket.nn#
This sub-module extends flax.linen with layers and tools that are useful to applications in quantum physics. Read more about the design goal of this module in their README
Linear Modules#
Implements a projection onto a symmetry group. |
|
A group convolution operation that is equivariant over a symmetry group. |
|
1D linear transformation module with mask for autoregressive NN. |
|
1D convolution module with mask for autoregressive NN. |
|
2D convolution module with mask for autoregressive NN. |
|
1D linear transformation module with mask for fast autoregressive NN. |
|
1D convolution module with mask for fast autoregressive NN. |
|
2D convolution module with mask for fast autoregressive NN. |
Activation functions#
Modifies a non-linearity to act separately on the real and imaginary parts |
|
Returns the relu non-linearity, applied separately to the real and imaginary parts |
|
selu applied separately to the real andimaginary parts of it's input. |
|
Logarithm of the hyperbolic cosine, implemented in a more stable way. |
|
Logarithm of the hyperbolic sine. |
|
Logarithm of the hyperbolic tangent. |
Miscellaneous Functions#
|
Encodes the array x into a set of binary-encoded variables described by the shape of a Hilbert space. |
|
Converts the configuration σ to a 64-bit integer denoting its index in the full Hilbert space. |
Utility functions#
|
Computes apply_fun(variables, states) on all states of hilbert and returns |
|
Blocks#
A Multi-Layer Perceptron with hidden layers. |
|
Implements the DeepSets architecture, which is permutation invariant and is suitable for the encoding of bosonic systems. |
|
A flax module symmetrizing the log-wavefunction \(\log\psi_\theta(\sigma)\) encoded into another flax module (flax.linen.Module) by summing over all possible symmetries \(g\) in a certain discrete permutation group \(G\). |