0.1.0

[mhmukadam]

What's Changed

• Add SO3 support by @luisenp in Add SO3 support #46
• Taoshaf.add so3 class by @fantaosha in Taoshaf.add so3 class #65
• Add SO3 rotate and unrotate by @fantaosha in Add SO3 rotate and unrotate #57
• add so3._hat_matrix_check() by @fantaosha in add so3._hat_matrix_check() #59
• Encapsulated data loading functions in tactile pushing example into a new class by @luisenp in Encapsulated data loading functions in tactile pushing example into a new class #51
• Added a TactilePoseEstimator class to easily create TheseusLayer for tactile pushing by @luisenp in Added a TactilePoseEstimator class to easily create TheseusLayer for tactile pushing #52
• Refactor tactile pushing model interface by @luisenp in Refactor tactile pushing model interface #55
• Minor fixes 02/01/2022 by @luisenp in Minor fixes 02/01/2022 #66
• add adjoint, hat and vee for SE3 by @fantaosha in add adjoint, hat and vee for SE3 #68
• Add SE3.exp_map() and SE3.log_map() by @fantaosha in Add SE3.exp_map() and SE3.log_map() #71
• Add SE3.compose() by @fantaosha in Add SE3.compose() #72
• add SE3.transform_from and SE3.transform_to by @fantaosha in add SE3.transform_from and SE3.transform_to #80
• Updated to CircleCI's next-gen images. by @luisenp in Updated to CircleCI's next-gen images. #89
• Updated README with libsuitesparse installation instructions. by @luisenp in Updated README with libsuitesparse installation instructions. #90
• Added kwarg to NonlinearOptimizer.optimizer() for tracking error history by @luisenp in Added kwarg to NonlinearOptimizer.optimizer() for tracking error history #82
• Merge infos results for truncated backward modes by @luisenp in Merge infos results for truncated backward modes #83
• Fix Issue 88 by @maurimo in Fix Issue 88 #97
• Fixed bug in Variable.update() that was breaking torch graph... by @luisenp in Fixed bug in Variable.update() that was breaking torch graph... #96
• Forced Gauss-Newton step for last iterations of truncated backward. by @luisenp in Forced Gauss-Newton step for last iterations of truncated backward. #81
• Add automatic differentiation on the Lie group tangent space by @fantaosha in Add automatic differentiation on the Lie group tangent space #74
• Add rand() to LieGroup by @fantaosha in Add rand() to LieGroup #95
• Fix SO2 rotate and unrotate jacobian by @fantaosha in Fix SO2 rotate and unrotate jacobian #58
• Add projection for sparse Jacobian matrices by @fantaosha in Add projection for sparse Jacobian matrices #98
• Add LieGroup Support for AutoDiffFunction by @fantaosha in Add LieGroup Support for AutoDiffFunction #99
• Add SE2.transform from() and fix shape bugs in SE2.transform_to() by @fantaosha in Add SE2.transform from() and fix shape bugs in SE2.transform_to() #103
• Switch SE3.transform_from and SE3.transform_to by @fantaosha in Switch SE3.transform_from and SE3.transform_to #104
• Enabled back ellipsoidal damping in LM with linear solvers support checks by @luisenp in Enabled back ellipsoidal damping in LM with linear solvers support checks #87
• Changed name of pytest mark for CUDA extensions. by @luisenp in Changed name of pytest mark for CUDA extensions. #102
• Fix backpropagation bugs in SO3 and SE3 log_map by @fantaosha in Fix backpropagation bugs in SO3 and SE3 log_map #109
• Add analytical jacobians for LieGroup.exp_map by @fantaosha in Add analytical jacobians for LieGroup.exp_map #110
• error and errorsquaredNorm optional data by @jeffin07 in error and errorsquaredNorm optional data #105
• Add analytical derivatives for LieGroup.log_map() by @fantaosha in Add analytical derivatives for LieGroup.log_map() #114
• Refactor SO3.to_quaternion to fix backward bugs and improve the accuracy around pi by @fantaosha in Refactor SO3.to_quaternion to fix backward bugs and improve the accuracy around pi #116
• Change RobotModel.forward_kinematics() interface by @luisenp in Change RobotModel.forward_kinematics() interface #94
• Added error handling for missing matplotlib and omegaconf installation when using theg by @luisenp in Added error handling for missing matplotlib and omegaconf installation when using theg #119
• Add jacobians argument to exp_map and log_map functions by @joeaortiz in Add jacobians argument to exp_map and log_map functions #122
• Add pose graph optimization example by @fantaosha in Add pose graph optimization example #118
• Added 'secret' option to keep constant step size when using truncated… by @luisenp in Added 'secret' option to keep constant step size when using truncated… #130
• Add jacobians for LieGroup.local() by @fantaosha in Add jacobians for LieGroup.local() #129
• Added method to update SE2 from x_y_theta. by @luisenp in Added method to update SE2 from x_y_theta. #131
• A complete version of Bundle Adjusment by @luisenp in A complete version of Bundle Adjusment  #117
• Fixed jacobians for Between and VariableDifference by @fantaosha in Fixed jacobians for Between and VariableDifference #133
• Added batching support to tactile pushing example. by @luisenp in Added batching support to tactile pushing example. #132
• Changed tactile pose example optim var initialization to use start pose for all vars by @luisenp in Changed tactile pose example optim var initialization to use start pose for all vars #137
• Override Vector operators in Point2 and Point3 by @jeffin07 in Override Vector operators in Point2 and Point3 #124
• Merge Between and VariableDiff with RelativePoseError and PosePriorError by @fantaosha in Merge Between and VariableDiff with RelativePoseError and PosePriorError #136
• Fix a bug in Objective.copy() by @fantaosha in Fix a bug in Objective.copy() #139
• Added option to force max iterations for TactilePoseEstimator. by @luisenp in Added option to force max iterations for TactilePoseEstimator. #141
• black version bump by @jeffin07 in black version bump #144
• Added code to split tactile pushing trajectories data into train/val by @luisenp in Added code to split tactile pushing trajectories data into train/val #143
• Removed RobotModel.dim() by @luisenp in Removed RobotModel.dim() #156
• [bug-fix] Fixed wrong data shape initialization for GPCostWeight.dt by @luisenp in [bug-fix] Fixed wrong data shape initialization for GPCostWeight.dt #157
• Made AutoDiffCostFunction._tmp_optim_vars copies of original by @luisenp in Made AutoDiffCostFunction._tmp_optim_vars copies of original #155
• Add forward kinematics using an URDF to theseus.embodied.kinematics. by @exhaustin in Add forward kinematics using an URDF to theseus.embodied.kinematics. #84
• Fixed dtype error in se3.py that came up in unit tests by @joeaortiz in Fixed dtype error in se3.py that came up in unit tests #158
• Add-ons for backward experiments on Tactile Pose Estimation by @luisenp in Add-ons for backward experiments on Tactile Pose Estimation #164
• Change unit tests to avoid making mypy a main requirement. by @luisenp in Change unit tests to avoid making mypy a main requirement. #168
• Update readme and contrib by @luisenp in Update readme and contrib #169
• Add ManifoldGaussian class for messages in belief propagation by @joeaortiz in Add ManifoldGaussian class for messages in belief propagation #121
• More efficient implementation of forward kinematics by @exhaustin in More efficient implementation of forward kinematics #175
• Changing setup virtualenv command. by @luisenp in Changing setup virtualenv command. #178
• Updated SDF object in collision cost functions whenever an aux var is updated by @luisenp in Updated SDF object in collision cost functions whenever an aux var is updated #177
• Fixed device bug that occurred when merging info in TRUNCATED backward modes by @luisenp in Fixed device bug that occurred when merging info in TRUNCATED backward modes #181
• Allow constant inputs to cost functions to be passed as floats by @jeffin07 in Allow constant inputs to cost functions to be passed as floats #150
• Minor changes to core test code. by @luisenp in Minor changes to core test code. #197
• adding logo by @mhmukadam in adding logo #200
• Added aliases for Difference and Between by @luisenp in Added aliases for Difference and Between #199
• Fixed infinite recursion in GPMotionModel.copy() by @luisenp in Fixed infinite recursion in GPMotionModel.copy() #201
• Fix bug in diagonal cost weight by @luisenp in Fix bug in diagonal cost weight #203
• Added a check in TheseusFunction that enforces copy() also copies the variables by @luisenp in Added a check in TheseusFunction that enforces copy() also copies the variables #202
• Fixed bug in Objective.error() that was updating data unncessarily. by @luisenp in Fixed bug in Objective.error() that was updating data unncessarily. #204
• DLM gradients by @rtqichen in DLM gradients #161
• Setup now uses torch for checking CUDA availability, and CI runs py3.9 tests by @luisenp in Setup now uses torch for checking CUDA availability, and CI runs py3.9 tests #206
• Update README to specify python versions and CUDA during install step + by @cpaxton in Update README to specify python versions and CUDA during install step + #207
• Vectorization refactor by @luisenp in Vectorization refactor #205
• Implement Robust Cost Function by @luisenp in Implement Robust Cost Function #148
• Added option for auto resetting LUCudaSparseSolver if the batch size needs to change by @luisenp in Added option for auto resetting LUCudaSparseSolver if the batch size needs to change #212
• Address comments on RobustCostFunction implementation by @luisenp in Address comments on RobustCostFunction implementation #213
• Moved the method that retracts all variables with a given delta to Objective by @luisenp in Moved the method that retracts all variables with a given delta to Objective #214
• Fixed flaky unit test for Collision2D jacobians. by @luisenp in Fixed flaky unit test for Collision2D jacobians. #216
• Bundle Adjustment using RobustCostFunction by @luisenp in Bundle Adjustment using RobustCostFunction #149
• Fixed the cross product bug for SE3.exp_map and SE3.log_map by @fantaosha in Fixed the cross product bug for SE3.exp_map and SE3.log_map #217
• Vectorize optimization variables retraction step by @luisenp in Vectorize optimization variables retraction step  #215
• Moved Vectorize(objective) to the Optimizer class. by @luisenp in Moved Vectorize(objective) to the Optimizer class. #218
• Added on-demand vectorization and also vectorized Objective.error(). by @luisenp in Added on-demand vectorization and also vectorized Objective.error(). #221
• Vectorize PGO by @luisenp in Vectorize PGO #211
• Destroy cusolver context in CusolverLUSolver destructor by @luisenp in Destroy cusolver context in CusolverLUSolver destructor #222
• Jacobian computation using loop by @fantaosha in Jacobian computation using loop  #225
• Vectorization handles singleton costs by @luisenp in Vectorization handles singleton costs #226
• Added close-formed jacobian for DLM perturbation cost. by @luisenp in Added close-formed jacobian for DLM perturbation cost. #224
• SE2/SE3/SO3 - consolidate EPS, add dtype-conditioned EPS and add float32 unit tests by @luisenp in SE2/SE3/SO3 - consolidate EPS, add dtype-conditioned EPS and add float32 unit tests #220
• Add normalization to Lie group by @fantaosha in Add normalization to Lie group #227
• Add normalization to Lie group constructor by @fantaosha in Add normalization to Lie group constructor #228
• Benchmarking PGO on main branch by @fantaosha in Benchmarking PGO on main branch #233
• Renamed Variable.data as Variable.tensor by @luisenp in Renamed Variable.data as Variable.tensor #229
• More data -> tensor renaming by @luisenp in More data -> tensor renaming #230
• Added state history tracking by @luisenp in Added state history tracking #234
• Unified all cost functions so that cost weight is the last non-default argument by @luisenp in Unified all cost functions so that cost weight is the last non-default argument #235
• Ensure that CHOLMOD python interface casts to 64-bit precision by @luisenp in Ensure that CHOLMOD python interface casts to 64-bit precision #238
• Add sphinx and readthedocs configuration by @luisenp in Add sphinx and readthedocs configuration #237
• Fixed bug in state history for matrix data tensors. by @luisenp in Fixed bug in state history for matrix data tensors. #240
• Added isort for examples folder by @luisenp in Added isort for examples folder #243
• Avoid batching SDF data, since it's shared by all trajectories. by @luisenp in Avoid batching SDF data, since it's shared by all trajectories. #246
• Fixed device bug in DLM perturbation's jacobians by @luisenp in Fixed device bug in DLM perturbation's jacobians #247
• Benchmark PGO on the main branch by @fantaosha in Benchmark PGO on the main branch #244
• Added checks to enforce 32- or 64-bit dtype by @luisenp in Added checks to enforce 32- or 64-bit dtype #245
• Update readme by @mhmukadam in Update readme #251
• Added MANIFEST.in and changed project name to theseus-ai. by @luisenp in Added MANIFEST.in and changed project name to theseus-ai. #252
• Adding simple example by @mhmukadam in Adding simple example #253
• Added option to clear cuda cache when vectorization cache is cleared. by @luisenp in Added option to clear cuda cache when vectorization cache is cleared. #249
• Added evaluation directory by @luisenp in Added evaluation directory #241

New Contributors

• @jeffin07 made their first contribution in error and errorsquaredNorm optional data #105
• @joeaortiz made their first contribution in Add jacobians argument to exp_map and log_map functions #122
• @exhaustin made their first contribution in Add forward kinematics using an URDF to theseus.embodied.kinematics. #84
• @rtqichen made their first contribution in DLM gradients #161
• @cpaxton made their first contribution in Update README to specify python versions and CUDA during install step + #207

Full Changelog: 0.1.0-b.2...0.1.0

---

This discussion was created from the release 0.1.0.