model#

Neural quantum states#

It’s recommended to use ResConv when one needs a deep NQS.

SingleDense(features, actfn, use_bias, ...)

Network with one dense layer \(\psi(s) = \prod f(W s + b)\).

RBM_Dense(features[, use_bias, dtype])

The restricted Boltzmann machine with one dense layer \(\psi(s) = \prod \cosh(W s + b)\).

SingleConv(channels, actfn[, use_bias, ...])

Network with one convolutional layer \(\psi(s) = \prod f(\mathrm{Conv}(s))\).

RBM_Conv(channels[, use_bias, dtype])

The restricted Boltzmann machine with one convolutional layer \(\psi(s) = \prod \cosh(\mathrm{Conv}(s))\).

ResConv(nblocks, channels, kernel_size, ...)

Deep convolutional residual network.

Fermionic Mean-field#

GeneralDet([U, dtype, out_dtype])

General determinant wavefunction \(\psi(n) = \mathrm{det}(n \star U)\).

RestrictedDet([U, dtype, out_dtype])

Restricted determinant wavefunction \(\psi(n) = \mathrm{det}(n_\uparrow \star U) \mathrm{det}(n_\downarrow \star U)\).

UnrestrictedDet([U, dtype, out_dtype])

Unrestricted determinant wavefunction \(\psi(n) = \mathrm{det}(n_\uparrow \star U_\uparrow) \mathrm{det}(n_\downarrow \star U_\downarrow)\).

MultiDet([ndets, U, coeffs, dtype, out_dtype])

Multi-determinant wavefunction \(\psi(n) = \sum_i c_i \mathrm{det}(n \star U_i)\).

GeneralPf([F, sublattice, dtype, out_dtype])

General Pfaffian wavefunction \(\psi(n) = \mathrm{pf}(n \star F \star n)\).

SingletPair([F, sublattice, dtype, out_dtype])

Singlet paired wavefunction \(\psi(n) = \mathrm{det}(n_\uparrow \star F \star n_\downarrow)\).

MultiPf([npfs, F, dtype, out_dtype])

Multi-Pfaffian wavefunction \(\psi(n) = \sum_i \mathrm{pf}(n \star F_i \star n)\).