Fastddm

Latest version: v0.3.12

Safety actively analyzes 682404 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 3 of 4

0.3

----

0.3.0

^^^^^^^^^^^^^^^^^^^

*Added*

* Image windowing function can now be used as an input for `ddm` function.
* Documentation (developer guidelines for contributors, installation, doc-strings, etc.)
* Noise estimators
* Intermediate scattering function
* Logo
* support for fixed expression in `fit_multik`
* general function `azimuthal_average_array` to perform an azimuthal average on a 3D `ndarray`

*Changed*

* `fit_multik` also returns the standard error of the parameters
* refactored functions for C++ & CUDA
* improved azimuthal average (changed behaviour of input bins variable)

*Fixed*

* mergesort & resample bugs with different dtypes
* initial fit at reference k value in `fit_multik` was not done properly
* CUDA compilation works well on various systems

.. *Deprecated*
.. *Removed*

0.2

----

0.2.0

^^^^^^^^^^^^^^^^^^^

*Added*

* ``ImageStructureFunction`` provides methods to retrieve the full plane representation from half-plane.
* ``SINGLE_PRECISION`` option at install time.
* Errors evaluated in ``AzimuthalAverage`` can be used in ``fit_multik``
* In ``fit_multik``, the user can now fix q-dependent parameter values via ``fixed_params``.
* In ``fit_multik``, the user can now fix q-dependent parameters range via ``fixed_params_min`` and ``fixed_params_max``.
* Results from ``fit_multik`` also include the ``k`` parameter for convenience.

*Changed*

* Now ``ImageStructureFunction`` data and err are stored using half-plane representation.
* Fit models are not saved to file due to incompatibilities with ``dill`` package.
* Updated pytest.

*Fixed*

* Passed parameters in ``fit_multik`` are not changed by the function.

.. *Deprecated*

*Removed*

* Removed unused functions and modules.

0.1.3

^^^^^^^^^^^^^^^^^^^

*Added*

* Reader function for raw image metadata (supports ``.tif`` and ``.nd2``)
* basic pytest routines
* ``chunkify`` function for time analysis (for non-stationary processes)
* variance in python backend
* uncertainty in azimuthal average (selectable via flag in ``azimuthal_average`` function)
* fit function to fit a model for multiple k/q values at once
* copyright information
* script to generate fit models for the intermediate scattering function and image structure function
* formatted (custom) binary file outputs (and readers/parsers) for the ``ImageStructureFunction`` and the ``AzimuthalAverage`` classes


*Changed*

* Now ``save`` method in ``ImageStructureFunction`` and ``AzimuthalAverage`` allow the usage of the filename without the argument keyword.
* ``read_images`` and related functions allow for the selection of color channels when reading image files.
* Improved speed in cuda fft2 step.
* moved from static to dynamic library compilation
* sped up the azimuthal average computation

*Fixed*

* fixed a bug in azimuthal average ``resample`` method
* fixed scaling issue in python backend fft mode
* fixed bug in variance calculation in cuda backend
* computing lag=0 in ddm python backend was causing an error due to faulty array slicing, this was fixed.

.. *Deprecated*
.. *Removed*

0.1.2

^^^^^^^^^^^^^^^^^^^

*Added*

* New ``power_spec`` and ``var`` properties in ``ImageStructureFunction`` and ``AzimuthalAverage`` classes, containing the average power spectrum of the input images and the variance of the temporal fluctuations of the 2D Fourier transformed images.
* ``len()`` method in ``ImageStructureFunction`` class, which returns the length of the underlying ``data``.

.. *Changed*

*Fixed*

* Fixed memory leak.
* Fixed evaluation of RAM memory in ``cuda`` core functions.

.. *Deprecated*
.. *Removed*

Page 3 of 4

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.