Matteo Ravasi
Matteo Ravasi
This PR handles https://github.com/PyLops/pylops/issues/278 and provides typing annotation to all functions and classes of pylops.
# Problem The `pylops.signalprocessing.FFT2D` operator crashes when trying to deal with cupy arrays. This seems due to the fact that `nfft` is stored as tuple - note that this was...
Some of the main linear operators may get a speed-up by simply adding `numba `_ ``@jit`` decorators to ``_matvec`` and ``_rmatvec`` methods. Others, such as ``FirstDerivative`` may be suited to...
# Motivation This Issue is intended to collect pointers to various transforms that we would like to include in PyLops (as well as discuss the level of maturity of the...
# Motivation The CG solver currently implements `x=A^-1b`. We may however also solver `x=(A+dampI)^-1b`. This is however currently not implemented. Note that `damp` is currently present as input values but...
# Motivation Following the discussion in https://github.com/PyLops/pylops/pull/363 we should consider also doing the same for linear operator # Solution Decorate `_matvect` and `_rmatvect` with `@abstractmethod`
## Feature Recently an Adaptive-FISTA algorithm was introduced in https://www.earthdoc.org/content/papers/10.3997/2214-4609.202229010. This algorithm shows faster convergence for problems where the model is expected to be very sparse (as it approximates the...
In the process of moving to v2.x we should consider the suggestion in https://github.com/PyLops/pylops/pull/336#issuecomment-1046220631 aimed at creating methods that can help users understand what has been deprecated and how it...
## Problem From version `1.8.0` scipy's `atol` and `btol` for `scipy.sparse.linalg.lsqr` have been changed from 1e-8 to 1e-6. Since we usually do not specify them in our test suite when...
## Feature The IRLS algorithm can handle L1 norm for the data misfit term as well as L1 norm for the model. However, in some scenarios users may want to...