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