Added functionality to find_peaks tool
Hi everybody First of all, this is my first post here, so excuse me if I do anything wrong. I'm mostly a user of SciPy, but recently I needed few functionalities from find_peaks that I discovered weren't implemented. I therefore decided to implement them myself. I'm now wondering if they could be useful for everybody and not just for me. Unfortunately I almost never worked on big projects on Git, so I'm a bit struggling understanding how should I proceed. First of all. The changes I'm talking of are: - Allow to work with unevenly spaced samples specifying a abscissas array x. More specifically: - Filter peaks depending on their distance with not-uniformly spaced points - Evaluate width and width's intersection points correctly with not-uniformly spaced points - Allow the above-mentioned abscissas to be a np.datetime64 array. In this case the minimum distance between two objects can (must) be specified as a np.timedelta64 value and the width of the peaks is also returned as an array of np.timedelta64 values. - Allow the output peaks to be sorted, either ascending or descending, depending either on their height or their prominence. - Allow to specify a maximum number of peaks to be returned. The output would then depend on the sorting method one has chosen. I would therefore ask you: - do you think these changes could be considered useful for the community? - could someone help me in my first-time submission process? Thank you very much and please tell me if I'm doing anything wrong Best regards Luca Amerio Luca
First two point seems interesting. Each of the next can be achieved in one line (with lexsort and fancy indexing) : it may be better to let that out of the function. I find these change interesting, but other opinion have to be considered. I can help you with the github workflow if you need so. Nicolas Cellier Post Doc Fellowship Université Savoie Mont-Blanc LAMA / LOCIE Le 23 nov. 2018 à 20:55, à 20:55, Luca <lucamerio89@gmail.com> a écrit:
Hi everybody
First of all, this is my first post here, so excuse me if I do anything wrong.
I'm mostly a user of SciPy, but recently I needed few functionalities from find_peaks that I discovered weren't implemented. I therefore decided to implement them myself. I'm now wondering if they could be useful for everybody and not just for me. Unfortunately I almost never worked on big projects on Git, so I'm a bit struggling understanding how should I proceed.
First of all. The changes I'm talking of are:
- Allow to work with unevenly spaced samples specifying a abscissas array x. More specifically: - Filter peaks depending on their distance with not-uniformly spaced points - Evaluate width and width's intersection points correctly with not-uniformly spaced points - Allow the above-mentioned abscissas to be a np.datetime64 array. In this case the minimum distance between two objects can (must) be specified as a np.timedelta64 value and the width of the peaks is also returned as an array of np.timedelta64 values. - Allow the output peaks to be sorted, either ascending or descending, depending either on their height or their prominence. - Allow to specify a maximum number of peaks to be returned. The output would then depend on the sorting method one has chosen.
I would therefore ask you:
- do you think these changes could be considered useful for the community? - could someone help me in my first-time submission process?
Thank you very much and please tell me if I'm doing anything wrong
Best regards Luca Amerio Luca
------------------------------------------------------------------------
_______________________________________________ SciPy-Dev mailing list SciPy-Dev@python.org https://mail.python.org/mailman/listinfo/scipy-dev
participants (3)
-
Lars Grüter -
Luca -
Nicolas Cellier