NIFTy merge requestshttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests2020-06-29T12:31:10Zhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/554Fix warning in doc generation2020-06-29T12:31:10ZPhilipp Arrasparras@mpa-garching.mpg.deFix warning in doc generationhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/545Proper constants2020-06-30T14:00:52ZPhilipp Arrasparras@mpa-garching.mpg.deProper constantsMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/555No flexibility or asperity for simple correlated field maker2020-06-30T21:24:51ZPhilipp Arrasparras@mpa-garching.mpg.deNo flexibility or asperity for simple correlated field maker@pfrank, Gordian had the idea that it might be worth while to be able to turn off the asperity and/or the fluctuations part of the correlated field model. For clarity's sake I started with the simple correlated field model. I am not clea...@pfrank, Gordian had the idea that it might be worth while to be able to turn off the asperity and/or the fluctuations part of the correlated field model. For clarity's sake I started with the simple correlated field model. I am not clear about how to disable the asperity part. It is probably easy for you. Do you have time to add this?
@gedenhofMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/550Try to avoid crashes in the line searcher2020-07-02T18:23:09ZMartin ReineckeTry to avoid crashes in the line searcherWith this patch, the line searcher checks for floating point errors, infinities and NaNs when extending its search interval. If it encounters one of those, it reduces the step size and tries again repeatedly. This actually seems to avoid...With this patch, the line searcher checks for floating point errors, infinities and NaNs when extending its search interval. If it encounters one of those, it reduces the step size and tries again repeatedly. This actually seems to avoid crashes in real-world applications.
@all: If you have problems with minimizations aborting due to overflow etc., please give this one a try and report whether or not it works for you!https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/557Paranoid linesearch2020-07-06T08:49:34ZPhilipp Arrasparras@mpa-garching.mpg.deParanoid linesearchhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/559Improvetesting2020-07-13T09:32:27ZPhilipp Arrasparras@mpa-garching.mpg.deImprovetestingMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/556Allow for opting out of asperity and flexibility2020-07-15T13:39:04ZGordian EdenhoferAllow for opting out of asperity and flexibilityhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/561Fixup `add_fluctuations` docstring2020-07-15T13:51:46ZLukas PlatzFixup `add_fluctuations` docstringSince !556 the docstring of `CorrelatedFieldMaker.add_fluctuations()` does not explain the meaning of the parameter tuples the function expects.
This patch mentions what kind of distributions the parameters control and what properties...Since !556 the docstring of `CorrelatedFieldMaker.add_fluctuations()` does not explain the meaning of the parameter tuples the function expects.
This patch mentions what kind of distributions the parameters control and what properties of them the parameters set.https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/558Amplitude -> Power spectrum2020-07-17T09:33:13ZPhilipp Arrasparras@mpa-garching.mpg.deAmplitude -> Power spectrum@reimar, @pfrank, @kjako, @mtr how do you like this change? The most important change is this one: https://gitlab.mpcdf.mpg.de/ift/nifty/-/commit/43761bb676411b83bc9786cde2f1e9cd3a8dc756#68e54f641853b0564eecbea76b4107d720683e0a_176_174
...@reimar, @pfrank, @kjako, @mtr how do you like this change? The most important change is this one: https://gitlab.mpcdf.mpg.de/ift/nifty/-/commit/43761bb676411b83bc9786cde2f1e9cd3a8dc756#68e54f641853b0564eecbea76b4107d720683e0a_176_174
It boils down to the fact that we model the actual power spectrum and not the amplitude with the integrated Wiener process. This saves one nonlinearity in the forward model and simplifies the interface, i.e. people do not need to remember to divide the values for `loglogavgslope` by two.https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/562Fix constant support of EnergyAdapter2020-07-18T13:33:02ZPhilipp Arrasparras@mpa-garching.mpg.deFix constant support of EnergyAdapter@pfrank does that work for you?
#311@pfrank does that work for you?
#311https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/563Update author list2020-07-18T13:39:40ZPhilipp Arrasparras@mpa-garching.mpg.deUpdate author listhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/564Cfm fix2020-08-12T15:14:33ZPhilipp Arrasparras@mpa-garching.mpg.deCfm fixMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/565Add softplus nonlinearity2020-08-24T15:35:33ZPhilipp Arrasparras@mpa-garching.mpg.deAdd softplus nonlinearityhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/566Import matplotlib2020-08-27T12:24:59ZPhilipp Arrasparras@mpa-garching.mpg.deImport matplotlibhttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/567fixup VariableCovarianceGaussianEnergy2020-09-02T13:17:13ZPhilipp Frankfixup VariableCovarianceGaussianEnergyThere was a small error in the metric of the `VariableCovarianceGaussianEnergy` Operator.
@parras @reimar @hutsch Please be aware of this as it changes the behaviour for real valued Gaussian likelihoods!There was a small error in the metric of the `VariableCovarianceGaussianEnergy` Operator.
@parras @reimar @hutsch Please be aware of this as it changes the behaviour for real valued Gaussian likelihoods!Philipp Arrasparras@mpa-garching.mpg.dePhilipp Arrasparras@mpa-garching.mpg.dehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/569Fix griddertests2020-09-17T13:18:58ZMartin ReineckeFix griddertestsI have updated the `ducc` docstrings to mention that the dirty image must have at least 32x32 pixels.
@parras, whet do you think?I have updated the `ducc` docstrings to mention that the dirty image must have at least 32x32 pixels.
@parras, whet do you think?https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/570Make Nonlinear operator test accept Operators2020-09-17T13:35:37ZReimar H LeikeMake Nonlinear operator test accept OperatorsThe Operator test did crash if it got an operator to test. This is obviously not the desired behavior.The Operator test did crash if it got an operator to test. This is obviously not the desired behavior.https://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/573Fix Sphinx warnings2020-10-20T11:12:59ZPhilipp Arrasparras@mpa-garching.mpg.deFix Sphinx warningsMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/572Fix Sphinx warnings2020-10-20T11:13:25ZPhilipp Arrasparras@mpa-garching.mpg.deFix Sphinx warningsMartin ReineckeMartin Reineckehttps://gitlab.mpcdf.mpg.de/ift/nifty/-/merge_requests/574Add helpful error message2020-10-26T14:18:07ZPhilipp Arrasparras@mpa-garching.mpg.deAdd helpful error messageMartin ReineckeMartin Reinecke