Sergio Sánchez Ramírez
Sergio Sánchez Ramírez
Currently we are using the default optimizer (`Greedy`) to get the contraction path of any `TensorNetwork`, but this optimizer... 1. ...can take some time to compute when in some cases...
All numerical methods are in the process to being moved to [bsc-quantic/Muscle.jl](https://github.com/bsc-quantic/Muscle.jl). Currently, it depends on OMEinsum too but has preliminary support for Reactant.jl and MLIR.jl. In the future, it...
⚠️ This is an idea I had that can be implemented in the future. We can experiment with it in other branch and decide on it in the future. Currently,...
Currently, `simple_update!` only supports 1 and 2-local gates. Generalizing to more sites shouldn't be very difficult, although it might more efficient to change the gate to a `MPO` representation.
This PR implements `LinearAlgebra.eigen` (and related methods, `eigvals` and `eigvecs`) for `Tensor`, and `LinearAlgebra.eigen!` for `AbstractTensorNetwork`. The design of this implementation is somewhat more similar to how Julia implements factorizations:...
Concretize `Ansatz` type with "lattice"/connectivity information and refactor subtypes on top of it
Inspired by [ITensorNetworks.jl](https://github.com/ITensor/ITensorNetworks.jl) (but not equal), I experimented with using a graph for representing the connectivity between different sites. In "ITensorNetworks.jl", they use the graph for representing the structure of...