Releases: ITensor/ITensor
Releases · ITensor/ITensor
Version v3.2.0
- Remove
format
in favor of tinyformat::format
, to avoid conflict with the new std::format
(#419).
Version v3.1.11
- [C++20] Replace
result_of
with invoke_result
(#400)
Version 3.1.10
- Julia compatible HDF5 Support for Complex Dense and Block Sparse (#398)
- Fixed issue with comparing
int
to size_t
in h5 codes (#390)
Version 3.1.9
This release adds support for writing and reading most ITensor objects to HDF5 files compatibly with the Julia version of ITensor. Complex-valued ITensors are not yet supported but will be soon.
- Add reading of MPS and MPO from HDF5 (#397)
- Add support for writing MPS and MPO to HDF5 (#396)
- HDF5 Support for QN Index and (Real) QN ITensor (#395)
- Add HDF5 support for QN (#394)
- HDF5 for dense real ITensors (#393)
- Add Checklists.txt for listing best practices
Version 3.1.8
- Fix missing return value MPO.svdBond method (thanks to Walter Hahn for recognizing this issue)
- Remove the make warnings on comparing size_t and int values (PR #385 Mingru Yang)
- Add syntax highlighting to readme (PR #387)
- Improvements to SVD backend options (PR #386 Daniel Bauernfeind)
- Update is_pod trait and improve .gitignore (PR #388 #391 Tanmoy B.)
v3.1.6
- Fix issue #311 regarding wrong fermion sign when no QNs conserved (PR #373)
- Add sample code for 2D Hubbard model with ky conservation
- Add citation info to README
v3.1.5
- Add CTMRG sample code.
- Update lapack wrapper functions to use LAPACK_INT (PR #369)
- Fix potential integer overflow in getContractedOffsets (PR #368)
- Reorganize and simplify TRG sample code
v3.1.4
- Switch headers to use pragma once (PR #356)
- Fix issue with colliding SVD indices (PR 357)
- More comprehensive README for sample codes
- Fix bug where LocalOp constructor errors if optional NumCenter variable left unspecified (PR #359)
- Fix issue with range function in case start > end (PR #362)
v3.1.3
- Remove util/print_macro.h from all.h (PR #352)
- Fix faulty QN MPS error check (PR #349)
- Improve LocalOp operator bool (PRs #351, #354 )
- Fix totalQN function to always succeed if lims set (PR #355)