v2.1.1

What's Changed

Fix:

• Fix bug in build_dataset.from_model. And update e3 example by @QG-phy in 5dfced2
• Fix: the bug in band plot. and update the argcheck for run.json and fix the type error for push overlap is None. @QG-phy at 041cd0f
• Add pyfiglet and update python version in pyproject.toml @QG-phy 58339f0

Example

• add: e3 example data by @floatingCatty in #203
• update ckpt ref by @floatingCatty in #204
• update silicon.ipynb example by @QG-phy at 3890ad7
• update checkpoints for silicon example by @QG-phy at b396783
• update band plot josn by @QG-phy at 62092ba
• update the train input.json for silicon @QG-phy da6c7da
• update data info.json remove the atomicdata option @QG-phy f2e7b64

Full Changelog: v2.1.0...v2.1.1

v2.1.0

What's Changed

• Fix: r_max set in e3 model by @floatingCatty in #197
• Update Dockerfile.main by @QG-phy in #198
• Qg phy patch 1 by @QG-phy in #199
• Update doc by @QG-phy in #201
• refactor(data preprocess): remove the cut-off options from info.json by @QG-phy in #200
• Update full options list by @QG-phy in #202

Full Changelog: v2.0.3...v2.1.0

v2.0.3

What's Changed

Feature

• Feature(nested tensor): support nested tensors for batchlize sampling of irregular shape data by @floatingCatty in #171
• feat:add support for molecular data. by @QG-phy in #172
• Feature(PyG data): support nested tensor's dtype and device transformation by @floatingCatty in #174
• add a new fermi level calculation method in band.py and its example by @AsymmetryChou in #176
• Feature: add support for Lmdb dataset by @floatingCatty in #177
• feat(loss): add new loss options to control number bands to exclude in in dft bands. by @QG-phy in #181
• Update: Organize e3, add e3 statistics in LMDB dataset and add e3 fitting of overlap by @floatingCatty in #183
• feat: add api to pybinding by @QG-phy in #186
• Update: support read multiple lmdb data file with LMDBDataset by @floatingCatty in #194

Fix:

• fix bug when dataset only has one atom and one frame by @AsymmetryChou in #175
• fix(deeptb.py): fix the bug raised when loading model from cuda and run on cpu by @floatingCatty in #178
• fix: the error occur when loading checkpoint from cuda to cpu or to cuda to continuous training by @floatingCatty in #182
• fix: delete checkpoint files exceeding maximum limit by @QG-phy in #184
• update lem.py, fix the natom error by @floatingCatty in #187
• Fix(tensor_product.py): fix the error when set radial embedding == False by @floatingCatty in #188
• Fix: fix frame index to compatible with 0 starts by @floatingCatty in #190
• Fix: h5 group blocks by @floatingCatty in #191
• Fix: CUDA availability check by @QG-phy in #193
• Update: support f/g/h orbital in sk hamiltonian by @floatingCatty in #192
• Fix: subtract e_fermi from blocks in ToPybinding by @QG-phy in #195
• build(deps): bump docker/build-push-action from 5 to 6 by @dependabot in #196

DOC

• update(doc): update document by @floatingCatty in #165
• update(doc): add basic api in doc by @floatingCatty in #166
• Update(doc): update deeptb-tbplas interfaces document by @floatingCatty in #167
• docs(example): deletes useless examples by @QG-phy in #169
• docs: update readme, easy install and contributing guide by @QG-phy in #173

Full Changelog: v2.0.2...v2.0.3

v2.0.2

What's Changed

Feature:

• feat:add support to pass kpoints np.array to get band eigenvalues. by @QG-phy in #145
• feat(data): add parse md trajectory of abacus by @floatingCatty in #144
• feat: add -v to get dptb version, and update the version.py to load version number and git version by @QG-phy in #158
• feat: update pyproject.toml and init.py to automatically get the version number. by @QG-phy in #160

Fix:

• Fix(nnsk): NNSK class in nnsk.py to use the get() method when accessing the full orbital by @floatingCatty in #139
• fix(data): fix the bug in dm parse by @floatingCatty in #142
• fix: idp(data) in nnsk and deeptb, and refactor soc switch in hr2hk by @floatingCatty in #152

Refactor:

• refactor: plotting code in dftb2nnsk.py fix push_decay method and update push options by @QG-phy in #156
• style: optmize the import of each submodule by @floatingCatty in #154
• Bump actions/setup-python from 4 to 5 by @dependabot in #131
• Bump actions/checkout from 3 to 4 by @dependabot in #130
• update yaml by @QG-phy in #164

Docs

• docs: Update dftb.md with hBN model training steps and fix formatting by @QG-phy in #157

Full Changelog: v2.0.1...v2.0.2

v2.0.1

What's Changed

Feature:

• feat(dftb): add support sk params from DFTB skf files . by @QG-phy in #133
• add E3 features (node/edge) to Hamiltonian/density blocks by @floatingCatty in #129
• feat(command): add a cskf command to collect the skfiles into a pth database by @QG-phy in #134
• update write block function in the command line by @floatingCatty in #132

Fix:

• fix(embedding): The numerical instability in constructing the descriptors for high symmetry structures. by @floatingCatty in #135
• fix(se2): update the smooth function in getting the env descriptor by @QG-phy in #136
• fix(run): update run.py, to make the dptb run command available: by @QG-phy in #128

Docs

• update mix.ep50.pth, mix.iter500.pth and train_config.json by @QG-phy in #126
• update h-BN example, including the checkpoints and the related doc. by @QG-phy in #127

Note: This version fixes numerical instability in constructing the environment descriptors in the env-correction term in deeptb for structures with high symmetry (usually for ideal crystal structures).

It is highly recommended for users to update to the new version.

Full Changelog: v2.0.0...v2.0.1

v2.0.0

What's Changed

• Mergev2 merge data nequip refactoring into main by @floatingCatty in #61
• pend run import, need to recover its function later by @floatingCatty in #63
• update carbon_hybrid model by @floatingCatty in #65
• Update README.md by @shishirkrpandey in #68
• Update docs for DeePTb v2.0 by @SharpLonde in #67
• Files fixed for various examples. by @SharpLonde in #69
• update dockerfile and toml file by @QG-phy in #71
• Dockerfile by @QG-phy in #73
• Dockerfile by @QG-phy in #74
• Dockerfile by @QG-phy in #75
• add fixed length on iter checkpoint by @floatingCatty in #76
• fix to align the checkpoint output index with the log output index by @floatingCatty in #77
• update unittest by @QG-phy in #78
• move 60 files update the data for test; move it to v1 by @QG-phy in #81
• Update dptb config command and its unit test by @QG-phy in #79
• add unit test for dptb bond command by @QG-phy in #82
• add unit test for train cmd by @QG-phy in #83
• Checkpoint files for examples by @SharpLonde in #84
• Merge changes during the E3 Hamiltonian model development into main by @floatingCatty in #85
• Test trainer by @QG-phy in #86
• add unit test and update some minor part by @QG-phy in #88
• add tests by @QG-phy in #89
• CI: Setup Dependabot for GitHub Actions by @njzjz-bot in #92
• add test by @QG-phy in #91
• Update unit_test.yml by @QG-phy in #98
• Bump actions/checkout from 3 to 4 by @dependabot in #95
• Update devcontainer.yml by @QG-phy in #99
• Bump docker/setup-buildx-action from 2 to 3 by @dependabot in #94
• Update unit_test.yml by @QG-phy in #100
• updata push option and saver by @QG-phy in #101
• Refactoring TBPLaS api to match the current code version by @floatingCatty in #102
• From v1json by @QG-phy in #103
• Bump docker/build-push-action from 4 to 5 by @dependabot in #96
• add parse and split options in data, and loss analysis by @floatingCatty in #90
• From v1json and To json and fix the support for NRL json by @QG-phy in #104
• refactor build api in data and nn by @floatingCatty in #106
• Loss analysis viz, debug tbplas api and support compute eigenvalues nk point each time by @floatingCatty in #111
• update dargs version by @QG-phy in #108
• update inputs for examples by @SharpLonde in #109
• Add run command by @QG-phy in #112
• Fix silicon by @floatingCatty in #113
• PR: Add support for SOC in the SKTB formula. by @QG-phy in #107
• Si demo by @QG-phy in #114
• CI: Mirror the repository to Gitee by @njzjz-bot in #117
• update DPTB to NNENV by @QG-phy in #115
• Create input_tags.md by @QG-phy in #118
• Update input_tags.rst by @QG-phy in #119
• Darg doc by @QG-phy in #120
• Darg doc by @QG-phy in #121
• Update dptb data by @SharpLonde in #116
• update doc and Sn soc example by @QG-phy in #122
• dev2main ready for release. by @QG-phy in #123
• Dev by @QG-phy in #125

New Contributors

• @shishirkrpandey made their first contribution in #68
• @SharpLonde made their first contribution in #67
• @njzjz-bot made their first contribution in #92
• @dependabot made their first contribution in #95

Full Changelog: v1.1.0...v2.0.0

v1.1.0

What's Changed

Fix the basic NEGF module

• Hartree2eV unit constant keep the same as dptb-negf by @AsymmetryChou in #58
• fix k-resolved transmission calculation bug and add kpoints sampling function, especially for NEGF by @AsymmetryChou in #59
• fix assert bug for k-points weight summation by @AsymmetryChou in #60

Full Changelog: v1.0.3...v1.1.0

v1.0.3

What's Changed

• hotfix: fix the validation in train nnsk and dptb, by @QG-phy in #54
• add doc and example for TBtrans interface by @AsymmetryChou in #52
• Example to get hr by @QG-phy in #56
• NRL update by @QG-phy in #57

Full Changelog: v1.0.2...v1.0.3

v1.0.2

What's Changed

• Hotfix remove the tracked output files from pytest by @QG-phy in #46
• update unit test for skint types and struct.proj_atom_symbols by @QG-phy in #47
• fix variable name bug due to rename by @AsymmetryChou in #49
• add TBtrans Interface tbtrans_init.py and its test files by @AsymmetryChou in #50

New Contributors

• @AsymmetryChou made their first contribution in #49

Full Changelog: v1.0.1...v1.0.2

v1.0.1

What's Changed

• Adding Non-equilibrium Green's Function module into postprocessing of DeePTB by @floatingCatty in #32
• Add Wannier TB for initial training SK TB parameters. by @QG-phy in #41

Full Changelog: v1.0.0...v1.0.1