Merged
Conversation
Rename rst files in `docs/content/api` and replace `SimPEG` for `simpeg` in `docs/content/api/index.rst`. Fixes #1437
Replace usage of `SimPEG` for `simpeg` in the Getting Started documentation pages.
Add checks to validate the `survey` and `lineID` inputs in the `convert_survey_3d_to_2d_lines` function. Add tests to check if errors are raised after passing invalid arguments. Improve the docstring of the function.
Always use the Pydata Sphinx theme for building the docs. Remove the `try` statement that used to check if the theme was installed and defaulted to the default theme if it wasn't.
Remove the `**kwargs` from the constructor of `UniformBackgroundField`. Make `receiver_list`, `amplitude`, `inclination` and `declination` as required arguments without any default value. Include type hints in the signature of the constructor. Add check for type of receivers to the constructor of `UniformBackgroundField`. Add test that checks the value of `UniformBackgroundField.b0`, passing `receiver_list` as an actual list of point receivers and as `None`. Add test that checks if the error is raised after passing invalid receiver type. Minor improvements to the docstring of the class.
Add a new issue template for opening maintenance issues.
Add instructions to the Contributing to SimPEG pages on how to update the environment once it's already created.
Since latest versions of `conda` make use of `libmamba` that achieves the same performance as `mamba` while installing packages and creating environments, we can stop recommending `mamba` to our users as a better alternative for `conda`. Update the admonitions in the instructions to install simpeg and to set up the environment for contributors. Mention the minimum version of `conda` that uses `libmamba` by default. Replace recommendations for Mambaforge to Miniforge.
Fix bug in the constructor of beta estimator directive classes: the `seed` argument was being passed as a positional argument to the parent's constructor, which would assign it to the `n_pw_iter` attribute instead to the `seed`, leading to unwanted behaviours. Pass the arguments to the constructor through keywords to make it more explicit and less prone to errors. Add tests that would fail without this bugfix. Remove the `n_pw_iter` and the `method` arguments from the constructor of `BaseBetaEstimator`.
Remove `**kwargs` from the constructor of the class, add `current` as one of the arguments. Improve the logic and error messages for the `location`, `location_a` and `location_b` arguments. Add tests for these new checks. Update docstring of the constructor to reflect latest changes. Fix the expected shape of the `location_a` and `location_b` arguments.
Replace the usage of legacy `numpy.random.seed()` in SimPEG codebase for the Numpy random number generator object that can be created through the `numpy.random.default_rng()` function. Add a new `typing` module that contain new type aliases, and create a new `RandomSeed` type alias for variables accepted by `numpy.random.default_rng()`. Extend the accepted values for `seed` arguments in several methods and functions to be `None` or `RandomSeed`. Apply these replacements to directives, `eigenvalue_by_power_iteration()`, `create_random_model()`, utility functions on `test_utils.py` and usage of `eigenvalue_by_power_iteration()` in tests. Update documentation and type hint of `make_synthetic_model()` allowing it to get a Numpy `Generator` as `random_seed`. Add a new documentation page for the new `typing` module.
Extend docstrings for the 3D field and simulation classes for FDEM and TDEM modules. Provide mathematical background for the implementations. Update format of docstrings following numpydoc style. Add a few constructor methods for some classes to fix some missing properties in child classes.
Add new implementation of the magnetic simulation using the integral formulation based on [Choclo](https://www.fatiando.org/choclo) and Numba. Allow users to run the simulation with Choclo as engine through a new `engine` argument of the constructor of the simulation class. Add `numba_parallel` argument to control if the simulation should be run in parallel or in a single thread. Move common methods and properties of the gravity and magnetic simulation to the base class for potential field simulations. Add new `potential_fields/_numba_utils.py` file with functions that are needed both in the magnetic and gravity Numba implementations. Refactor and extend tests for the magnetic simulation, including tests that assert the accuracy of the new implementation against analytic solutions. Add new tests for the base class for potential field simulations. Update examples using Choclo as engine, include admonitions instructing how to use it. --------- Co-authored-by: Jacob Edman <jacob.edman@koboldmetals.com> Co-authored-by: Joseph Capriotti <josephrcapriotti@gmail.com>
Fix minor typos in EM docstrings.
Update call of `_print_verbose_msg_init_end` in `GaussianMixtureWithPrior`: since `scikit-learn>=1.5.0` it requires two positional arguments. Use a `try-except` block to maintain support for older versions of `scikit-learn`. Create a new private method for the `GaussianMixtureWithPrior` class to wrap the `_print_verbose_msg_init_end` method including the `try-except` block. Add tests. Fix #1475
Add a version switcher to the navbar of the docs, using the built in one in PyData Sphinx theme. Point to the experimental `versions.json` file located in `doctest.simpeg.xyz` that should serve the files in [simpeg/simpeg-doctest](https://github.com/simpeg/simpeg-doctest). Add experimental staged to Azure Pipelines that pushes docs to the `simpeg-doctest` repository. One is triggered nightly and deploys the latest docs in `main` to the `dev` branch in `simpeg-doctest` and updates the `dev` submodule in `gh-pages`. The other one is triggered after a release and pushes the new version of the docs to `gh-pages` branch in `simpeg-doctest` onto a new folder, while also updating the `latest` link.
Make use of the `random_seed` argument of the `make_synthetic_data` method in magnetic tests. Remove the lines that set a global `np.random.seed` in those tests. Part of the solution to #1289
Update the `linkcode_resolve` function in Sphinx configuration file so the links to the source code of each function, class or method points to the actual code corresponding to that version, and not always to the code in `main`. Update some Azure configurations: checkout simpeg repo without shallow depth and fetching tags. Clean up the working directory before installing SimPEG so the version doesn't have an extra hash. --------- Co-authored-by: Joseph Capriotti <josephrcapriotti@gmail.com>
Fix issues in the new (experimental) scripts for deployment of docs to `simpeg-doctest`. Fetch the `gh-pages` before trying to checkout it. This is needed because we clone the repo with `--depth 1`. When passing this option, git appends the `--single-branch` option meaning the `dev` branch is the only branch that is being cloned. Improve how we clone the `simpeg-doctest` repo: clone quietly to reduce output lines, and specify the branch while deploying release docs (in case the default branch in the repo gets changed).
Include the version of SimPEG being used to run a simulation in the inversion log. Fixes #1474
Replace the usage of the deprecated functions in `numpy.random` module for the Numpy's random number generator class and its methods, in most of the FDEM tests. Increase number of iterations for checking derivatives where needed. Part of the solution to #1289 --------- Co-authored-by: Lindsey Heagy <lindseyheagy@gmail.com>
Add a new `random_seed` argument to the `test()` method of objective functions to control their random state. Use Numpy's random number generator for managing the random state and the generation of random numbers. Minor improvements to the implementation of the test methods. Update tests that make use of these methods, and make them to use a seed in every case. Part of the solution to #1289
Add `random_seed` argument to `IdentityMap.test`. Add a new `random_seed` argument to the `test` method of `IdentityMap` to control the random state for creating the `m` array if it's not provided. Update tests to use the `random_seed`. If the `m` array is passed while running the `test()` method, specify the keyword for the argument to avoid relying only on its position.
Replace the usage of the deprecated functions in `numpy.random` module for the Numpy's random number generator class and its methods, in most of the TDEM tests. Part of the solution to #1289
Make use of the `random_seed` arguments in a few objective function tests that were missed in previous PR. Related to #1448.
Correcting contour colors in a gravity plot in the User Guide for a gravity plot: correctly use the color limits both in the contours and while building the custom color bar.
Configure Sphinx autodoc to hide type hints from signatures of functions, methods and classes to improve readability. Expected types for each parameter should be listed in the docstring.
Remove the `simpeg/maps.py` file and create a new `simpeg/maps` directory. Add private files for each category of map classes inside the new submodule. Make these classes public through the `simpeg/maps/__init__.py`.
Moves simpeg to a pyproject.toml based setup. * as much configuration as possible in `pyproject.toml` * general repository cleanup * ci shell scripts
Create a custom RST landing page using `sphinx-design` that includes cards to the main sections of the docs. Move some content from the old landing page (present in the `README.rst`) into new sections in the User Guide: `citing.rst` and `about-simpeg.rst`.
Add a new page to the How to Guide that explains how to move a mesh to a given survey area. This is handy when working with real-world data, whose coordinates are going to be given in projected plain coordinates.
Add an admonition pointing users to the respective tutorials in User Tutorials.
Apply some minor improvements to the docs of `UpdateSensitivityWeights`: add admonition, fix usage of the `class` directive, update class names mentioned in the docs, improve list of options for different arguments.
Update the iteration write out order. For the longest time, the iteration printers have written out mismatched values at each iteration, and also did not print the last value of the inversion. This updates the write outs to happen after the line-search (matching the behavior of the save directives), but before the directives themselves alter the objective function. I've also removed several un-needed notifications: * setting bfgsH0 * only set on BFGS subclasses and * if the UpdatePreconditioner directive is not in the directive list (It handles the appoxHinv). * Removes the repeated notifications about no nans in the model, and instead raises an error if the model vector has any nans. Writers: * The iteration writer formatting logic was rather convoluted to me, so I refactored it a bit to make it a little easier to understand what it was doing...
Fix wrong B field when using the magnetic dipole source with the H-J formulation. When using the magnetic vector potential to define the source term in the H-J formulation, multiply by the face integration matrix before taking the curl and multiply by the inverse edge integration matrix afterwards.
Bump minimum required version of Python to 3.11.
Add a new page to the docs that specify the guidelines we follow to support older versions of Python and Numpy.
Remove most of the DC tutorials, and add an admonition pointing users to the respective tutorials in User Tutorials. Part of #1555
Improve the assertions in dipole source tests for EM. Compute the root mean square of the difference between the numerical and analytic fields in FDEM, and check it's below a certain tolerance. Use `np.testing.assert_allclose` in TDEM tests to compare the predicted data for both the B and H formulations.
…CG minimizers (#1703) #### Summary Updates the minimizer constructors used in exampled, tutorials, and tests to no longer use the deprecated arguments related to conjugate gradient minimizers. #### What does this implement/fix? <!--Please explain your changes.--> #### Additional information After passing build, we'll want to check how well the examples & tutorials that use `ProjectedGNCG` match previous results. As these now use a relative tolerance on CG vs. an absolute tolerance. I suspect they will not change much as most of them were terminated by the `cg_maxiter` condition previously. --------- Co-authored-by: Santiago Soler <santisoler@fastmail.com>
Make the `ComplexMap.deriv` to return a sparse diagonal matrix instead of a `LinearOperator`. This makes the method to match the signature of other mappings. Add tests that check the expected behaviour. --------- Co-authored-by: Joseph Capriotti <josephrcapriotti@gmail.com> Co-authored-by: Santiago Soler <santisoler@fastmail.com>
Standardize signature of `deriv` method across mappings: make sure they are all defined as `deriv(m, v=None)`. Implement the dot product between the derivative matrix and `v` if `v` is not `None`. Add tests for the modified methods. --------- Co-authored-by: Santiago Soler <santisoler@fastmail.com>
Update how mappings are being used in regularizations: regularize on `mapping(m) - mapping(m_ref)` instead of `mapping(m - m_ref)`. Update implementations and documentation of `Smallness`, `SmoothnessFirstOrder`, `SmoothnessSecondOrder`, and `SmoothnessFullGradient`: the `f_m` will compute difference of the mapped model and reference model, while the `f_m_deriv` will not make use of the reference model now, since it's not carried out in the derivative. Add tests to check this intended behaviour.
#### Fixes for pymatsolver 0.4.0 <!-- Add a summary of this Pull Request --> #### PR Checklist * [x] If this is a work in progress PR, set as a Draft PR * [x] Linted my code according to the [style guides](https://docs.simpeg.xyz/latest/content/getting_started/contributing/code-style.html). * [x] Added [tests](https://docs.simpeg.xyz/latest/content/getting_started/contributing/testing.html) to verify changes to the code. * [x] Added necessary documentation to any new functions/classes following the expect [style](https://docs.simpeg.xyz/latest/content/getting_started/contributing/documentation.html). * [x] Marked as ready for review (if this is was a draft PR), and converted to a Pull Request * [x] Tagged ``@simpeg/simpeg-developers`` when ready for review. #### What does this implement/fix? Squeeze these two locations in the SimulationDC operation (to be consistent on output of `__inner_mat_mul `operation) #### Additional information The `__inner_mat_mul` function will "squeeze" out single length second axes, so this fix causes it to be consistent for this operation in CG. This function had to perform this way because of the way pymatsoler previously handled single length array inputs.
Remove the duplicated current being multiplied when computing the `Mejs` in `LineCurrent`. Add a test that checks that the electric fields are linear in the current.
Apply some fixes to LaTeX equations in some regularization classes.
Make the `get_indices_block` function to always return an array of integers for the cells that belong to the specified block, instead of returning a tuple with just a single array. Improve its implementation. Fix it for 1D mesh since it wasn't working properly. Improve error messages: raise better errors than just `AssertError`s and improve messages. Avoid overwriting the `p0` and `p1` arrays in-place if they are not provided in the right order (west-south-bottom and then east-north-top). Add tests for the function. Update usage of the function across tests and examples. This introduces a backward incompatible change.
Remove the `Data.index_dictionary` property. Remove the `gtg_diagonal` property from gravity simulation. Remove the `components` property from gravity and magnetic surveys. Remove the `HasModel.deleteTheseOnModelUpdate` property.
Add new `shift_to_discrete_topography` function that shifts locations relative to discrete surface topography. Add also a new `get_discrete_topography` function that generates discrete topography locations out of a mesh and its active cells. We measure electric fields at the Earth's surface when performing MT surveys. Like DC/IP, the original measurement locations of the electric fields can end up in air cells when we define discrete surface topography. These functions allow the user to shift locations relative to discrete topography on Tensor and Tree meshes. For Airborne NSEM, they also allow the user to preserve the original flight heights. Deprecate the following functions: `gettopoCC`, `drapeTopotoLoc`, and `closestPointsGrid`. --------- Co-authored-by: Santiago Soler <santisoler@fastmail.com>
Deprecate the unused `force` and `topography` arguments of the method. Add tests to check the warnings are raised.
Fix a few LaTeX equations in `simpeg.potential_fields.magnetics.Simulation3DDifferential` that weren't rendering correctly.
Implement the B field computation from a closed-loop wire source in the TDEM code. We first compute the vector potential using the Biot-Savart Law (implemented in geoana) and then take the curl to get the initial B-field. Adds the computation of B with a step-off waveform using the B or H formulations. --------- Co-authored-by: Joseph Capriotti <josephrcapriotti@gmail.com> Co-authored-by: Santiago Soler <santisoler@fastmail.com>
Add changelog for next release of SimPEG: v0.25.0. Add the new version to the version switcher. --------- Co-authored-by: Joseph Capriotti <josephrcapriotti@gmail.com> Co-authored-by: Lindsey Heagy <lindseyheagy@gmail.com>
Forgot to add the new release notes to the index, so they can show up in the website. Update the release date. Minor corrections to RST syntax.
Fix the name of the `threshold_value` argument in the docstring of the `UpdateSensitivityWeights` directive.
jedman
approved these changes
Dec 4, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
13 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bring in the latest updates from simpeg.