search

cheminfo / common-spectrum

https://cheminfo.github.io/common-spectrum/.
MIT License
3 stars 2 forks source link

manual peak picking with `optimize=true` should also return width #63

Open kjappelbaum opened 3 years ago

kjappelbaum commented 3 years ago

i think a more natural behavior of optimize=true would be to also fit a shape to make it consistent with the automatic peak picking. This is also important for many techniques such as XPS or XRD where we actually also want to know the width of the peak. I.e., i would call https://github.com/mljs/global-spectral-deconvolution/blob/985d9aa4d539c2ccf5afd84c51b49d58037da5ac/src/post/optimizePeaks.js#L23 and not simply look for the closest min/max

what do you think @lpatiny

lpatiny commented 3 years ago

Currently we give back the width after optimizePeaks: https://github.com/mljs/global-spectral-deconvolution/blob/81f286b6cd6fb54e48a25731a95e20c3ec739b6e/src/post/__tests__/optimizePeaks.test.js#L21-L29

But I guess we would need to provide also the other optimization parameters if there are some. @kjappelbaum what would you like to have ? The width in the manual peak picking result ? I fully agree.

kjappelbaum commented 3 years ago

The width in the manual peak picking result ? I fully agree.

yes, i'd like to have the width in the manual peak picking to have an output that is interoperable with the one from automatic picking. I'll try to copy the option interface we use for the automatic peak picking.