adaptive

adaptive copied to clipboard

Promote AveragingLearner to LinearFitLearner

2

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/18)) _opened by Anton Akhmerov ([@anton-akhmerov](https://gitlab.kwant-project.org/anton-akhmerov)) at 2017-09-06T09:31:46.601Z_ A more advanced version of averaging would be performing a linear fit. The logic here should be largely...

jbweston

Mark latest points in the live plot

6

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/24)) _opened by Anton Akhmerov ([@anton-akhmerov](https://gitlab.kwant-project.org/anton-akhmerov)) at 2017-10-01T10:31:03.988Z_ The live plot is a useful tool in judging the quality of the result, and figuring out what...

jbweston

CQuad does not order actions based on error estimates

1

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/28)) _opened by Joseph Weston ([@jbweston](https://gitlab.kwant-project.org/jbweston)) at 2017-10-27T14:02:20.393Z_ The current implementation (IIRC) is meant to replicate the behaviour of Christoph's code in the case of sequential...

jbweston

Blocking behavior of the runner

7

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/31)) _opened by Anton Akhmerov ([@anton-akhmerov](https://gitlab.kwant-project.org/anton-akhmerov)) at 2017-11-13T14:39:40.396Z_ Right now the runner is stopped when the user launches anything that blocks the kernel, and that is...

jbweston

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/40)) _opened by Anton Akhmerov ([@anton-akhmerov](https://gitlab.kwant-project.org/anton-akhmerov)) at 2017-12-22T13:15:54.380Z_ If the loss is related to the error from the true value (e.g. the error of the integral...

basnijholt

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/41)) _opened by Bas Nijholt ([@basnijholt](https://gitlab.kwant-project.org/basnijholt)) at 2018-01-02T11:43:09.087Z_ Since efcccfcc41373932da9a87109e0085d4cc65b7a1 we set the error of the very first interval to a very large number (largest Python...

basnijholt

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/42)) _opened by Bas Nijholt ([@basnijholt](https://gitlab.kwant-project.org/basnijholt)) at 2018-01-02T12:30:21.790Z_ In d357a64eb2df8bc7a99c2ecfdc7cf0bebcb8822c we change `IntegratorLearner.ivals` from a `SortedSet` (sorted by `err`) to a normal `set` because the `SortedSet`...

basnijholt

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/43)) _opened by Bas Nijholt ([@basnijholt](https://gitlab.kwant-project.org/basnijholt)) at 2018-01-05T10:55:36.350Z_ Currently, when an interval is removed because the numerical precision of the integral (or error) cannot be improved...

basnijholt

Suggestion: simplify learner interface

10

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/45)) _opened by Christoph Groth ([@cwg](https://gitlab.kwant-project.org/cwg)) at 2018-01-20T10:03:09.612Z_ It seems to me that "adaptive" could greatly profit by becoming trivially extensible. The adaptive package consists of:...

basnijholt

Suggestion: hide the actual function to be evaluated from learners

8

## ([original issue on GitLab](https://gitlab.kwant-project.org/qt/adaptive/issues/46)) _opened by Christoph Groth ([@cwg](https://gitlab.kwant-project.org/cwg)) at 2018-01-20T14:43:17.989Z_ Learners do not need to have access to the function object they are trying to predict. I believe...

basnijholt