netket.nn.to_array#
- netket.nn.to_array(hilbert, apply_fun, variables, *, normalize=True, allgather=True, chunk_size=None)[source]#
- Computes apply_fun(variables, states) on all states of hilbert and returns
the results as a vector.
- Parameters:
normalize (
bool) – If True, the vector is normalized to have L2-norm 1.allgather (
bool) –- When running with MPI:
If True, the final wave function is stored in full at all MPI ranks.
- When running with netket_experimental_sharding=True:
If allgather=True, the final wave function is a fully replicated array If allgather=False, the final wave function is a sharded array, padded with zeros to the next multiple of the number of devices
chunk_size (
Optional[int]) – Optional integer to specify the largest chunks of samples that the model will be evaluated upon. By default it is None, and when specified samples are split into chunks of at most chunk_size.hilbert (DiscreteHilbert)
apply_fun (Callable[[Any, ndarray | Array], ndarray | Array])
variables (Any)
- Return type:
Returns: