v0.1.0

Releasing v0.1.0, it is equivalent to v0.0.28

v0.0.28

What's new? 🎉🎉

add coffee dataset

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.27

What's new? 🎉🎉

• PointScaler: Scale your spectra by the intensity value given at a certain index or wavenumber! This substitutes the old IndexScaler, as it extends its functionality
• SelectFeautes: An advanced feature selector compare to Range Cut. It allows you to choose any range of indices or wavenumbers (continuous or discontinuous) and select the features

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.26

What's new? 🎉🎉

• MinMaxScaler: change functionality, not it will subtract the min and divide by the difference between the min and the max. If the parameter use_min is False, then it will just divide by the max.

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.25

What's new? 🎉🎉

Improvements ✨✨

Bug fixes 🐛🐛

• RangeCut now has a different input order: start, end and wavenumber (optional). Optional inputs are defined at the end. start and end index are found after fitting the method and not upon instantiation. This is because in scikitlearn, instanciation attributes cannot be modified.

• ConstantCorrection: Same changes as RangeCut

v0.0.24

What's new? 🎉🎉

• Include fermentation datasets 🔝

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.23

What's new? 🎉🎉

• Add extended multiplicative scatter correction
• Add Robust Normal variate!

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.22

What's new? 🎉🎉

• Add docstrings in all available methods

Improvements ✨✨

Bug fixes 🐛🐛

v0.0.21

What's new? 🎉🎉

Improvements ✨✨

• Increase performance of ArPls() by a factor of x40

Bug fixes 🐛🐛

v0.0.20

What's new? 🎉🎉

• New preprocessing added for baseline correction. It is called assymetrically reweighted penalized least squares (ArPls()). The current implementation in based on the following work:

Sung-June Baek a, Aaron Park *a, Young-Jin Ahn a and Jaebum Choo, Baseline correction using asymmetrically reweighted penalized least squares smoothing

Improvements ✨✨

Bug fixes 🐛🐛