NIFTy merge requestshttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests2019-04-09T13:50:05Zhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/310Single plot and cosmetics2019-04-09T13:50:05ZLukas PlatzSingle plot and cosmetics`ScalingOperator`: Replace magic integer constants with readable constants from parent class `LinearOperator`.
`sugar.single_plot`: function that creates a plot with just one call. No more boilerplate while plotting, yay!`ScalingOperator`: Replace magic integer constants with readable constants from parent class `LinearOperator`.
`sugar.single_plot`: function that creates a plot with just one call. No more boilerplate while plotting, yay!https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/309Linear amplitude model2019-04-08T11:37:18ZReimar H LeikeLinear amplitude modelExporting the linear part of the amplitude model. Also deleted the requirement that Gaussian energies cannot have MultiDomain as domainExporting the linear part of the amplitude model. Also deleted the requirement that Gaussian energies cannot have MultiDomain as domainMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/308fixing metric of sum operator2019-04-03T12:09:04ZReimar H Leikefixing metric of sum operatormetric of sum operator was failing due to domain mismatch when there it is chained with a more complicated operator.metric of sum operator was failing due to domain mismatch when there it is chained with a more complicated operator.Philipp Arrasparras@mpa-garching.mpg.dePhilipp Arrasparras@mpa-garching.mpg.dehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/307Fix handling of metric in _OpSum2019-03-15T16:01:54ZPhilipp Arrasparras@mpa-garching.mpg.deFix handling of metric in _OpSumReady to be mergedReady to be mergedhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/306Generalize do_adjust_variances2019-04-19T08:58:11ZPhilipp Arrasparras@mpa-garching.mpg.deGeneralize do_adjust_variancesThe adjust variances function is unnecessarily specific at the moment. This is now generalized to the case d=A*xi where xi is an apriori white Gaussian random field and A is a general operator.
Do we want to do some renamings? Is there ...The adjust variances function is unnecessarily specific at the moment. This is now generalized to the case d=A*xi where xi is an apriori white Gaussian random field and A is a general operator.
Do we want to do some renamings? Is there perhaps a better name than `do_adjust_variances`?https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/305fixes data_objects.numpy_do.absmax2019-03-13T17:52:29ZLukas Platzfixes data_objects.numpy_do.absmaxhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/304Add NFFTOperator2019-03-13T14:51:33ZPhilipp Arrasparras@mpa-garching.mpg.deAdd NFFTOperatorhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/303Documentation feedback2019-04-10T08:07:59ZPhilipp Arrasparras@mpa-garching.mpg.deDocumentation feedbackhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/302Sampling from scaling2019-03-07T14:09:50ZPhilipp Arrasparras@mpa-garching.mpg.deSampling from scalinghttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/301parallelization for mirrored KL2019-08-16T05:52:39ZReimar H Leikeparallelization for mirrored KLThe KL with mirror_samples=True can now be parallelized by another factor of 2. Threads will reuse drawn samples to mirror them if possible.The KL with mirror_samples=True can now be parallelized by another factor of 2. Threads will reuse drawn samples to mirror them if possible.Martin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/300remove obsolete division methods2019-02-28T14:02:21ZMartin Reineckeremove obsolete division methodshttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/299Convolution improvements2019-02-27T15:10:30ZMartin ReineckeConvolution improvementsThis branch adds two features to the existing convolutions:
- convolution is now also supported on RGSpaces (tested)
- convolution can now be applied to a subspace of a DomainTuple (not yet tested)
@lplatz, @mawandro, please give this a...This branch adds two features to the existing convolutions:
- convolution is now also supported on RGSpaces (tested)
- convolution can now be applied to a subspace of a DomainTuple (not yet tested)
@lplatz, @mawandro, please give this a try!https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/298re-introduce mpi_experiments2019-02-27T15:11:31ZMartin Reineckere-introduce mpi_experimentshttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/297Spherical Convolution operator2019-02-19T13:34:55ZLukas PlatzSpherical Convolution operatorHi Martin,
I used your `apply_spherical_convolution` function as a template to create both
- a function to create convolution kernels from functions within `LMSpace`s: `get_conv_kernel_from_func`
- two operators that perform convolution...Hi Martin,
I used your `apply_spherical_convolution` function as a template to create both
- a function to create convolution kernels from functions within `LMSpace`s: `get_conv_kernel_from_func`
- two operators that perform convolutions on spherical fields, one taking a kernel directly , one creating it from a function (`SphericalConvolutionOperator`, `SphericalFuncConvolutionOperator`)
I tried to enable git-praise attribution of the code snippets to you by first committing your function under your name, but it does not seem to have worked.
Could you please take a look at the code?
Is desirable to force the input domain to be a DomainTuple?
I also added a small demo, which seems to work fine.
Thanks!Martin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/296more fixes2019-02-14T14:30:51ZMartin Reineckemore fixeshttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/295Simplify for const2019-03-04T13:52:21ZPhilipp Arrasparras@mpa-garching.mpg.deSimplify for consthttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/294typo2019-02-14T12:35:34ZMartin Reinecketypohttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/293now clipper takes fields or multifields2019-02-15T12:55:37ZReimar H Leikenow clipper takes fields or multifieldsShould solve #260 .Should solve #260 .Martin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/292fix installation command2019-02-07T22:12:14ZMartin Reineckefix installation commandhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/291arguments need to be reversed2019-02-05T14:23:17ZReimar H Leikearguments need to be reversedFound a small bug. But there seems to be a larger issue with dtype left, my code throws an error inside the fft. Maybe I do a minimal example later.Found a small bug. But there seems to be a larger issue with dtype left, my code throws an error inside the fft. Maybe I do a minimal example later.Martin ReineckeMartin Reinecke