Tiny bug fix for v1.9.0 (dependencies, nothing to do with the actual code).

• Simulation

  • New attribute select, which returns a reduced simulation containing the selected sources, receivers, and frequencies.
  • New cleaning option 'gradient' (Simulation.clean('gradient')), cleaning only gradient-related values.

• Maintenance

  • Testing: added Python 3.13, 3.14; removed Python 3.10.

• Maintenance:

  • Replace scipy.interpolate.interpnd with scipy.interpolate._interpnd (should be a temporary solution, would be better to not use private modules).
  • Fix test failures; they were related to the change of SciPy v1.15 of constants.mu_0 value changed to the more precise double precision value (scipy/scipy#11345).

• Maintenance:

  • Restricting to SciPy<1.15. The function scipy.interpolate.interpnd is deprecated, and scipy.interpolate.interpnd._ndim_coords_from_array was removed. Needs fixing in emg3d.maps._points_from_grids.
  • Re-adding a test for normalizing source warning; avoid rounding over and over again in recursion of creating a dipole/wire source.

• Bugfix and maintenance:

  • Fix bug in Dipole/Wire sources when coordinates are mixed increasing and decreasing over several cells #348.
  • Changed build backend from setup.py to pyproject.toml.

The code is now compatible with NumPy v2.

• Created foundation for new module inversion.

• Bumped the minimum requirements to:

  • Python 3.10
  • SciPy 1.10
  • empymod 2.3.2
  • Numba (without minimum version)

• Maintenance

  • Testing: dropped Python 3.9 (Python 3.13 not added yet).
  • Update for NumPy v2: mainly np.infty -> np.inf.
  • Add notes for ipympl (interactive plots in modern Jupyter).
  • Reduce code by making use of new SciPy new features (complex-valued map_coordinate; lazy loading).

• New hard dependency: empymod changed from being a soft dependency to being a hard dependency.

• Simulation

  • A new parameter tol_gradient can be provided in the dict solver_opts; by default it is set to the value of tol. tol is the value used for compute (the forward), tol_gradient is used for gradient/jtvec and jvec (the gradient). In inversions, one can set the tolerance for the gradient often to a lower value, which saves computation time.

• Survey

  • New attributes isfinite and finite_data. Former returns the indices of the finite data, latter returns directly the finite data.

• Electrodes

  • Fixed TxMagneticDipole-representation and improved documentation of the magnetic sources.

• Maintenance:

  • Bumped the minimum requirements to:

    • Python 3.9
    • SciPy 1.9
    • Numba 0.53
    • empymod 2.3.0 (NEW requirement)

  • Testing: added Python 3.12, dropped Python 3.8.

  • Fix remaining outdated python setup.py commands.

  • Many small things to keep the package updated.

• io: New function convert to convert a file that was saved with emg3d from one file format to another file format.

• Installation through pip has new the option pip install emg3d[full] which installs all soft dependencies as well.

• Bumped the minimum requirements to:

  • Python 3.8
  • NumPy 1.19
  • SciPy 1.5
  • Numba 0.50

• Bug fixes, small improvements and maintenance

  • Testing: added Python 3.11, dropped Python 3.7.
  • Adjust copyright notice to only include original year, so it has not to be adjusted each year.

• Small improvements to the ellipse-function (and the CLI):

  • The major and minor axis of the ellipse are new forced to be at least 1e-9.
  • A radius of 0.0 provided through the CLI is now respected (was before
    overwritten).

• empymod is more verbose when using layered computations.

• Maintenance:

  • Replace deprecated sphinx-panels with sphinx-design.

The simulation class takes new the parameters layered and layered_opts,
where the default values are False and None, respectively. If layered=True,
there will be no 3D computations. Instead, it will create a local layered (1D)
model for each source-receiver pair, and compute the response using the
semi-analytical code empymod (which needs to be installed manually, as it
is a soft dependency). In this case an eventual gradient is computed using the
finite-difference method, not the adjoint-state method, perturbing each layer
slightly. The main purpose of these layered computations is for quick checks,
QC, verifications, etc. Layered computation is also possible through the CLI,
through the new flag -l or --layered, and a new section [layered]
in the config file.

Other changes (many of them related to the above):

• Model instances have a new attribute exctract_1d, which returns a layered
  (1D) model, extracted from the 3D model according the provided parameters;
  see :attr:emg3d.models.Model.extract_1d.

• CLI takes new the boolean add_noise in the section [noise_opts]
  (default is True).

• Maps: New function ellipse_indices returning a boolean indicating which
  points fall within a general ellipse for the provided input parameters.

• Bug fixes, small improvements and maintenance

  • Simulation.misfit returns an ndarray again instead of an DataArray (was
    erroneously changed in v1.2.1).
  • Write json can now handle NumPy int/bool/float.
  • A clean on a Simulation now removes correctly the weights.
  • Capture error in jtvec if weight is complex NaN (should be real).
  • Model: mapping can now be an already instantiated map (sort of
    undocumented).
  • Cleaned-up the namespace by setting dir() explicitly.
  • Replace pytest-flake8 by plain flake8.
  • Moved all multiprocessing-related functions to _multiprocessing.py.