Preprocessing

Sometimes we need to preprocess the data prior to analyze it. The modules in this category deal with this problem.

Missing data

When the observations contains missing data, it is necessary to reconstruct the invalid information before processing it further. Here you can learn more about this procedure.

Smoothing

If the functional data observations are noisy, smoothing the data allows a better representation of the true underlying functions. You can learn more about the smoothing methods provided by scikit-fda here.

Registration

Sometimes, the functional data may be misaligned, or the phase variation should be ignored in the analysis. To align the data and eliminate the phase variation, we need to use registration methods. Here you can learn more about the registration methods available in the library.

Dimensionality Reduction

The functional data may have too many features so we cannot analyse the data with clarity. To better understand the data, we need to use dimensionality reduction methods that can reduce the number of features while still preserving the most relevant information. Here you can learn more about the dimension reduction methods available in the library.

Feature construction

When dealing with functional data we might want to construct new features that can be used as additional inputs to the machine learning algorithms. The expectation is that these features make explicit characteristics that facilitate the learning process. To construct new features from the curves, feature construction methods are available. Here you can learn more about the feature construction methods available in the library.