v2.5.1

What's Changed

• bump jax and jaxlib versions to 0.4.* by @tylerflex in #1326
• Fix to mesher error in very specific cases by @momchil-flex in #1325
• add string matching to assert_log_level by @tylerflex in #1327
• Added validation for symmetry along zero size dimension by @tomflexcompute in #1344
• Warn if time monitor start time is after simulation run time by @momchil-flex in #1350
• Unified validation check for dependency fields by @dbochkov-flexcompute in #1348
• changelog for unified validation check for missing dependency fields by @dbochkov-flexcompute in #1354
• Update black version by @momchil-flex in #1356

Full Changelog: v2.5.0...v2.5.1

v2.5.0

What's Changed

• tidy3d stub by @magiWei in #1108
• Avoid creating unecessary geometry groups of single elements by @lucas-flexcompute in #1177
• time zone in webapi log by @tylerflex in #1183
• merge current state of develop (2.4.2) into pre/2.5 by @tylerflex in #1185
• Heat solver interface by @dbochkov-flexcompute in #1083
• add divergence mention to PML warning by @tylerflex in #1195
• Issue a warning when accessing unavailable group index data (#1169) by @lucas-flexcompute in #1178
• Prep 2.5.0rc1 by @dbochkov-flexcompute in #1198
• broadband adjoint (2.5) by @tylerflex in #1192
• Add to_gds to Simulation, Structure and Geometry (#883) by @lucas-flexcompute in #1194
• Optimize workbench url. by @magiWei in #1206
• Adding verbose argument to estimate_cost and real_cost by @momchil-flex in #1207
• unpin boto and click by @tylerflex in #1210
• 2.5 changes by @tylerflex in #1190
• Space-time modulation of all types of materials by @weiliangjin2021 in #1188
• Bugfix for polarization sorting with non-existent modes by @dbochkov-flexcompute in #1215
• Add support for geometry transformations by @lucas-flexcompute in #1203
• handle case with no adjoint sources by @tylerflex in #1218
• Setting protocol_version to None if solver_version provided by @momchil-flex in #1219
• Scene/AbstractSimualtion/Simulation/HeatSimulation cleaning by @dbochkov-flexcompute in #1211
• Error instead of warning for zero-dim boundaries not periodic or Bloch by @shashwat-sh in #1221
• Avoid duplicate log messages in Jupyter by @lucas-flexcompute in #1225
• Truncate circular filter in edge case by @tylerflex in #1227
• Fix bug in eps_model for integer frequency arrays by @caseyflex in #1230
• Removing indent in string stored in hdf5 by @momchil-flex in #1228
• Added two-photon absorption nonlinearity by @caseyflex in #1172
• use mode solver fields in conversion methods by @tylerflex in #1236
• proper equality checking between tidy3d base model by @tylerflex in #1237
• Split ModeSolver.data_raw into smaller methods by @momchil-flex in #1233
• Added PECMedium to IsotropicUniformMediumType, and added PEC2D by @caseyflex in #1193
• Added LO-TO form of PoleResidue model by @caseyflex in #1240
• Correctly handle 2D/1D custom medium when exporting to GDS by @lucas-flexcompute in #1247
• SourceTime.plot_spectrum missing return after refactor by @momchil-flex in #1250
• adds SimulationData.mnt_data_from_file() method to load individual MonitorData by @tylerflex in #1249
• Removing special direction='-' handling in adjoint sources after solv… by @momchil-flex in #1252
• better error message when user try to download data that doesn't exist. by @magiWei in #1246
• Fix for file names containing dots by @lucas-flexcompute in #1253
• matplotlib >= 3.5 by @tylerflex in #1260
• Fix Component Modeler Batch Path by @tylerflex in #1259
• add ComponentModeler.plot_sim_eps() by @tylerflex in #1269
• Fix for 2D PEC by @shashwat-sh in #1256
• Allow downsampling of near fields for field projection monitors by @shashwat-sh in #1272
• apply phase to FieldData and SimulationData.plot_field by @tylerflex in #1271
• Ensuring mode solver fields are returned in requested precision by @momchil-flex in #1266
• Enforcing limits on the internal memory needed by monitors by @momchil-flex in #1273
• static jax structures & dispersive adjoint by @tylerflex in #1281
• adjoint warn nonlinear by @tylerflex in #1283
• Some private helper properties in ModeSolver by @momchil-flex in #1255
• Unstructured grid datasets by @dbochkov-flexcompute in #1251
• Remove heat warning by @dbochkov-flexcompute in #1289
• Spatial filtering of far fields by windowing near fields by @shashwat-sh in #1275
• Fix to fully anisotropic eps_model when an array of freqs is provided by @momchil-flex in #1290
• Error if any frequencies are smaller than 1e5 Hz by @momchil-flex in #1284
• Field projection validators by @shashwat-sh in #1288
• Making bend radius a positive float by @momchil-flex in #1298
• Replacing np.isclose with math.isclose in transformed validator by @momchil-flex in #1296
• On the fly vtk import by @dbochkov-flexcompute in #1300
• Return 0 spectrum if no times are above the dft cutoff by @momchil-flex in #1297
• Rename Silicon in material library by @weiliangjin2021 in #1295
• Allow field projections through a manually-specified medium by @shashwat-sh in #1291
• Validate that ModeSolver.plane intersects ModeSolver.simulation by @momchil-flex in #1306
• Allow bend_radius to be negative, but not close to zero by @momchil-flex in #1307
• Change definition of FluxTime and DFT fields with complex fields by @caseyflex in #1279
• Convenience property in mode solver to get number of cells, freqs, an… by @momchil-flex in #1305
• Print cost breakdown in estimate_cost if there is mode or post-procesing charge by @momchil-flex in #1309
• Mode solver eps_complex check at central frequency only by @momchil-flex in #1312
• limit vtk version by @dbochkov-flexcompute in #1313
• Added max_num_poles at least min_num_poles validator to FastDispersionFitter by @caseyflex in #1314
• 2.5 branch by @tylerflex in #1170

Full Changelog: v2.4.3...v2.5.0

v2.5.0rc3

What's Changed

• Avoid duplicate log messages in Jupyter by @lucas-flexcompute in #1225
• Truncate circular filter in edge case by @tylerflex in #1227
• Fix bug in eps_model for integer frequency arrays by @caseyflex in #1230
• Removing indent in string stored in hdf5 by @momchil-flex in #1228
• Added two-photon absorption nonlinearity by @caseyflex in #1172
• use mode solver fields in conversion methods by @tylerflex in #1236
• proper equality checking between tidy3d base model by @tylerflex in #1237
• Split ModeSolver.data_raw into smaller methods by @momchil-flex in #1233
• Added PECMedium to IsotropicUniformMediumType, and added PEC2D by @caseyflex in #1193
• Added LO-TO form of PoleResidue model by @caseyflex in #1240
• Correctly handle 2D/1D custom medium when exporting to GDS by @lucas-flexcompute in #1247
• SourceTime.plot_spectrum missing return after refactor by @momchil-flex in #1250
• adds SimulationData.mnt_data_from_file() method to load individual MonitorData by @tylerflex in #1249
• Removing special direction='-' handling in adjoint sources after solv… by @momchil-flex in #1252
• better error message when user try to download data that doesn't exist. by @magiWei in #1246
• Fix for file names containing dots by @lucas-flexcompute in #1253
• matplotlib >= 3.5 by @tylerflex in #1260
• Fix Component Modeler Batch Path by @tylerflex in #1259
• add ComponentModeler.plot_sim_eps() by @tylerflex in #1269
• Fix for 2D PEC by @shashwat-sh in #1256
• Allow downsampling of near fields for field projection monitors by @shashwat-sh in #1272
• apply phase to FieldData and SimulationData.plot_field by @tylerflex in #1271
• Ensuring mode solver fields are returned in requested precision by @momchil-flex in #1266
• Enforcing limits on the internal memory needed by monitors by @momchil-flex in #1273

Full Changelog: v2.5.0rc2...v2.5.0rc3

v2.5.0rc2

What's Changed

• Limit number of sources to 1000 by @momchil-flex in #1202
• Test initialization speed of a metalens with 10_000 structures by @momchil-flex in #1201
• Fix updater for hdf5 files with custom data, and add hdf5 files to automatic updater testing by @momchil-flex in #1200
• broadband adjoint (2.5) by @tylerflex in #1192
• Add to_gds to Simulation, Structure and Geometry (#883) by @lucas-flexcompute in #1194
• Take make_json_compatible out of _json_string as a standalone method by @momchil-flex in #1204
• Optimize workbench url. by @magiWei in #1206
• Adding verbose argument to estimate_cost and real_cost by @momchil-flex in #1207
• unpin boto and click by @tylerflex in #1210
• 2.5 changes by @tylerflex in #1190
• Space-time modulation of all types of materials by @weiliangjin2021 in #1188
• Bugfix for polarization sorting with non-existent modes by @dbochkov-flexcompute in #1215
• Add support for geometry transformations by @lucas-flexcompute in #1203
• handle case with no adjoint sources by @tylerflex in #1218
• Setting protocol_version to None if solver_version provided by @momchil-flex in #1219
• Scene/AbstractSimualtion/Simulation/HeatSimulation cleaning by @dbochkov-flexcompute in #1211
• Error instead of warning for zero-dim boundaries not periodic or Bloch by @shashwat-sh in #1221

Full Changelog: v2.5.0rc1...v2.5.0rc2

v2.4.3

What's Changed

• fix bug with freqs as np.ndarray in output monitors by @tylerflex in #1189
• lsf - added more monitor parsing capability by @alec-flexcompute in #1184
• better description for TimeMonitor.interval by @tylerflex in #1187
• Limit number of sources to 1000 by @momchil-flex in #1202
• Test initialization speed of a metalens with 10_000 structures by @momchil-flex in #1201
• Fix updater for hdf5 files with custom data, and add hdf5 files to automatic updater testing by @momchil-flex in #1200
• Take make_json_compatible out of _json_string as a standalone method by @momchil-flex in #1204

Full Changelog: v2.4.2...v2.4.3

v2.5.0rc1

What's Changed

• tidy3d stub by @magiWei in #1108
• Avoid creating unecessary geometry groups of single elements by @lucas-flexcompute in #1177
• fix bug with freqs as np.ndarray in output monitors by @tylerflex in #1189
• lsf - added more monitor parsing capability by @alec-flexcompute in #1184
• better description for TimeMonitor.interval by @tylerflex in #1187
• time zone in webapi log by @tylerflex in #1183
• merge current state of develop (2.4.2) into pre/2.5 by @tylerflex in #1185
• Heat solver interface by @dbochkov-flexcompute in #1083
• add divergence mention to PML warning by @tylerflex in #1195
• Issue a warning when accessing unavailable group index data (#1169) by @lucas-flexcompute in #1178
• Prep 2.5.0rc1 by @dbochkov-flexcompute in #1198

Full Changelog: v2.4.2...v2.5.0rc1

v2.4.2

changelog and schema for 2.4.2

v2.4.1

ready for 2.4.1

v2.4.0

What's Changed

• Add option to control log suppression by @lucas-flexcompute in #1009
• Fixed import and plotting of trimesh.Scene by @caseyflex in #1018
• supports abort task by @magiWei in #998
• add test for mode solver to and from hdf5 with custom medium by @tylerflex in #1020
• feat(SCEM-3114): plot3d little polish by @majinge7846 in #1032
• filtering, projection, penalization, and optimization convenience wrappers for adjoint plugin by @tylerflex in #1012
• Mode Solver convenience methods for simulation export by @alec-flexcompute in #1033
• Added fast dispersion fitter by @caseyflex in #977
• fix Environment object does not attribute _current if TIDY3D_ENV system variable is not empty by @JunchengXue in #1027
• calculate simulation monitor data size by @magiWei in #1034
• Added CustomSourceTime by @caseyflex in #994
• remove redundant phase from shifted source in smatrix by @tylerflex in #1036
• fix DeviceArray import errors by @tylerflex in #1038
• Nyquist step also taking frequency-domain monitors into account by @weiliangjin2021 in #1037
• Medium perturbation by @dbochkov-flexcompute in #996
• Alec/intersections 2dbox by @alec-flexcompute in #1040
• fix trianglemesh custom_dataset bug by @tylerflex in #1041
• setting version to 2.4.0rc2 by @weiliangjin2021 in #1048
• Update material_reference.py by @dominauta in #1045
• Improve docstrings for Gaussian beam sources to clarify the location of the waist by @shashwat-sh in #1043
• added bounding capabilities for plot and plot_eps by @alec-flexcompute in #1044
• Changed CustomSourceTime to automatically modulate at freq0 by @caseyflex in #1051
• Option to allow DC component in GaussianPulse spectrum by @caseyflex in #1054
• Use pytest's tmp path instead of hard-codded ones by @lucas-flexcompute in #1058
• install jax normally on windows by @tylerflex in #1059
• add python3.11 to tests by @tylerflex in #1060
• Fix to mode solver with negative angle_theta by @momchil-flex in #1055
• Library updated using the fast fitter and inclusion of Palik materials by @e-g-melo in #1047
• include input_structures in jax_sim_data.plot_field() by @tylerflex in #1067
• numerically stable radius penalty by @tylerflex in #1063
• 2d error intersections_2dbox by @alec-flexcompute in #1061
• Added NonlinearSusceptibility with chi3 support by @caseyflex in #992
• Updating monitors to colocate to primal grid, and colocate=True by default by @momchil-flex in #851
• Add SetOperation to construct more complex geometry by @lucas-flexcompute in #1052
• Lucas/geometry refactor by @lucas-flexcompute in #1064
• JEPT-279: reafactor the tidy3d python client to support new mode solver data table by @yuweiflexcompute in #1029
• supports upload task with gz by @magiWei in #999
• Fully anisotropic docstring fixes by @dbochkov-flexcompute in #1074
• Rich console width by @dbochkov-flexcompute in #1070
• upgrade to pydantic v2 by @tylerflex in #1073
• Equalize vertex processing in Geometry and PolySlab by @lucas-flexcompute in #1075
• Update apodization field description in PermittivityMonitor by @weiliangjin2021 in #1077
• Remove warning from CustomSourceTime by @caseyflex in #1078
• cast abs(JaxDataArray) to jnp.ndarray by @tylerflex in #1079
• fix 2d intersection plotting in Simulation by @tylerflex in #1080
• Fix backwards compatibility break by @lucas-flexcompute in #1081
• fix lineplot zorder by @tylerflex in #1085
• Lucas/recursive geometry fixes by @lucas-flexcompute in #1084
• Dispersion fitter test coverage improvement by @weiliangjin2021 in #1089
• add uat2 env by @magiWei in #1091
• SCEM-3421: create folder if folder does not exist by @JunchengXue in #1092
• Fix bug in simulation epsilon bounds min always set to 1 by @momchil-flex in #1095
• Extended lower limit of frequency range for graphene by @caseyflex in #1098
• Final status update in web.monitor with the correct field decay by @momchil-flex in #1102
• Fixed rendering of equation in 'NonlinearSusceptibility' class. by @caseyflex in #1105
• Improved warnings with Medium2D by @caseyflex in #1097
• Remote mode solver now supports double by @weiliangjin2021 in #1106
• adapt to AWS lambda environment by @magiWei in #1100
• Make TIDY3D_DIR configurable via environment variable by @yaugenst in #1132
• Frequency lower bound and target_neff shift in mode solver by @weiliangjin2021 in #1130
• adjoint filter cleanup and additions by @tylerflex in #1093
• ComponentModeler freqs as FreqArray by @momchil-flex in #1103
• Moved warning for 2d geometry without Medium2D to Simulation.structures by @caseyflex in #1137
• Changed source spectrum calculation to work for large number of time steps by @caseyflex in #1134
• add a local adjoint speed test by @tylerflex in #1138
• Fix warning capture by @dbochkov-flexcompute in #1136
• Reorganizing grid correction application by @momchil-flex in #1139
• add client source by @magiWei in #1142
• Validate group_index_step not too small when using single precision by @weiliangjin2021 in #1145
• set colocate=False in output field monitors for backwards compatibility by @tylerflex in #1144
• add explanation about feature size vs filter radius by @tylerflex in #1149
• Fixes for group_index_step validator (#1143) by @lucas-flexcompute in #1152
• Branch for 2.4 release by @momchil-flex in #1006
• Fixing inf bounds handling in PolySlab by @weiliangjin2021 in #1150

New Contributors

• @majinge7846 made their first contribution in #1032
• @alec-flexcompute made their first contribution in #1033
• @dominauta made their first contribution in #1045
• @e-g-melo made their first contribution in #1047
• @yuweiflexcompute made their first contribution in #1029
• @yaugenst made their first contribution in #1132

Full Changelog: v2.3.2...v2.4.0

v2.4.0rc2

What's Changed

• fix DeviceArray import errors by @tylerflex in #1038
• Nyquist step also taking frequency-domain monitors into account by @weiliangjin2021 in #1037
• Medium perturbation by @dbochkov-flexcompute in #996
• Alec/intersections 2dbox by @alec-flexcompute in #1040
• fix trianglemesh custom_dataset bug by @tylerflex in #1041
• setting version to 2.4.0rc2 by @weiliangjin2021 in #1048
• Update material_reference.py by @dominauta in #1045
• Improve docstrings for Gaussian beam sources to clarify the location of the waist by @shashwat-sh in #1043
• added bounding capabilities for plot and plot_eps by @alec-flexcompute in #1044
• Changed CustomSourceTime to automatically modulate at freq0 by @caseyflex in #1051
• Option to allow DC component in GaussianPulse spectrum by @caseyflex in #1054
• Use pytest's tmp path instead of hard-codded ones by @lucas-flexcompute in #1058
• install jax normally on windows by @tylerflex in #1059
• add python3.11 to tests by @tylerflex in #1060
• Fix to mode solver with negative angle_theta by @momchil-flex in #1055
• Library updated using the fast fitter and inclusion of Palik materials by @e-g-melo in #1047
• include input_structures in jax_sim_data.plot_field() by @tylerflex in #1067
• numerically stable radius penalty by @tylerflex in #1063
• 2d error intersections_2dbox by @alec-flexcompute in #1061
• Added NonlinearSusceptibility with chi3 support by @caseyflex in #992
• Updating monitors to colocate to primal grid, and colocate=True by default by @momchil-flex in #851
• Add SetOperation to construct more complex geometry by @lucas-flexcompute in #1052
• Lucas/geometry refactor by @lucas-flexcompute in #1064
• JEPT-279: reafactor the tidy3d python client to support new mode solver data table by @yuweiflexcompute in #1029
• supports upload task with gz by @magiWei in #999
• Fully anisotropic docstring fixes by @dbochkov-flexcompute in #1074
• Rich console width by @dbochkov-flexcompute in #1070
• upgrade to pydantic v2 by @tylerflex in #1073
• Equalize vertex processing in Geometry and PolySlab by @lucas-flexcompute in #1075
• Update apodization field description in PermittivityMonitor by @weiliangjin2021 in #1077
• Remove warning from CustomSourceTime by @caseyflex in #1078
• cast abs(JaxDataArray) to jnp.ndarray by @tylerflex in #1079
• fix 2d intersection plotting in Simulation by @tylerflex in #1080
• Fix backwards compatibility break by @lucas-flexcompute in #1081
• fix lineplot zorder by @tylerflex in #1085
• Lucas/recursive geometry fixes by @lucas-flexcompute in #1084

New Contributors

• @dominauta made their first contribution in #1045
• @e-g-melo made their first contribution in #1047
• @yuweiflexcompute made their first contribution in #1029

Full Changelog: v2.4.0rc1...v2.4.0rc2