|
3 | 3 | Changelog |
4 | 4 | ========= |
5 | 5 |
|
| 6 | +Version 2.5.0 |
| 7 | +------------- |
| 8 | + |
| 9 | +*Released on: 21/06/2025* |
| 10 | + |
| 11 | +* Added `cuda` engine to :py:class:`pylops.waveeqprocessing.Kirchhoff` |
| 12 | + operator |
| 13 | +* Added `Opbasis` and `optimal_coeff` to |
| 14 | + :py:class:`pylops.optimization.cls_sparsity.OMP` |
| 15 | +* Added `solver` to the input parameters of the `_oneshot` |
| 16 | + internal methods of :py:class:`pylops.waveeqprocessing.AcousticWave2D` |
| 17 | + to avoid recreating it for every shot |
| 18 | +* Added `kwargs_fft` to `pylops.signalprocessing.FFT`, |
| 19 | + :py:class:`pylops.signalprocessing.FFT2D`, and |
| 20 | + :py:class:`pylops.signalprocessing.FFTND` |
| 21 | +* Fix bug in :py:func:`pylops.waveeqprocessing.MDD` when using |
| 22 | + CuPy arrays for `G` and `d` with `twosided=True` and `add_negative=True` |
| 23 | +* Fix bug in :py:class:`pylops.signalprocessing.FourierRadon3D` |
| 24 | + in the default choice of `num_threads_per_blocks` |
| 25 | +* Fix bug in :py:class:`pylops.signalprocessing.Convolve1D` |
| 26 | + in the definition of `pad` and `padd` when applying the |
| 27 | + operator to a CuPy array |
| 28 | +* Fix bug in :py:class:`pylops.optimization.cls_sparsity.OMP` avoiding |
| 29 | + passing `explicit` in the creation of `_ColumnLinearOperator` |
| 30 | +* Fix bug in :py:class:`pylops.optimization.cls_sparsity.OMP` callback |
| 31 | + method as `cols` was not passed not allowing ``x`` to be |
| 32 | + properly reconstructed |
| 33 | +* Fix bug in :py:func:`pylops.waveeqprocessing.SeismicInterpolation` |
| 34 | + in calculation of `sampling` when not passed |
| 35 | + |
| 36 | + |
6 | 37 | Version 2.4.0 |
7 | 38 | ------------- |
8 | 39 |
|
@@ -189,7 +220,7 @@ Version 1.18.0 |
189 | 220 | * Extended :py:func:`pylops.Laplacian` to N-dimensional arrays |
190 | 221 | * Added `forward` kind to :py:class:`pylops.SecondDerivative` and |
191 | 222 | :py:func:`pylops.Laplacian` |
192 | | -* Added `chirp-sliding` kind to :py:class:`pylops.waveeqprocessing.seismicinterpolation.SeismicInterpolation` |
| 223 | +* Added `chirp-sliding` kind to :py:func:`pylops.waveeqprocessing.SeismicInterpolation` |
193 | 224 | * Fixed bug due to the new internal structure of `LinearOperator` submodule introduced in `scipy1.8.0` |
194 | 225 |
|
195 | 226 |
|
@@ -389,7 +420,7 @@ Version 1.9.1 |
389 | 420 | :py:class:`pylops.signalprocessing.FFT2D` to ensure that the type of the input |
390 | 421 | vector is retained when applying forward and adjoint. |
391 | 422 | * Added ``dtype`` parameter to the ``FFT`` calls in the definition of the |
392 | | - :py:class:`pylops.waveeqprocessing.MDD` operation. This ensure that the type |
| 423 | + :py:func:`pylops.waveeqprocessing.MDD` operation. This ensure that the type |
393 | 424 | of the real part of ``G`` input is enforced to the output vectors of the |
394 | 425 | forward and adjoint operations. |
395 | 426 |
|
|
0 commit comments