Tame the Curse of Dimensionality! Learn Dimensionality Reduction (PCA) and implement it with Python and Scikit-Learn.Image supply: unsplash.com.In the novel Flatland, characters dwelling in a two-dimensional world discover themselves perplexed and unable to grasp after they encounter a three-dimensional being. I take advantage of this analogy as an example how comparable phenomena happen in Machine Learning when coping with issues involving 1000’s and even tens of millions of dimensions (i.e. options): shocking phenomena occur, which have disastrous implications on our Machine Learning fashions.I’m positive you felt surprised, at the least as soon as, by the large variety of options concerned in fashionable Machine Learning issues. Every Data Science practitioner, eventually, will face this problem. This article will discover the theoretical foundations and the Python implementation of probably the most used Dimensionality Reduction algorithm: Principal Component Analysis (PCA).Why do we have to scale back the variety of options?Datasets involving 1000’s and even tens of millions of options are frequent these days. Adding new options to a dataset can usher in priceless data, nevertheless, they are going to sluggish the coaching course of and make it tougher to seek out good patterns and options. In Data Science that is referred to as the Curse of Dimensionality and it usually results in skewed interpretation of information and inaccurate predictions.Machine studying practitioners like us can profit from the truth that for many ML issues, the variety of options might be diminished constantly. For instance, take into account an image: the pixels close to the border usually don’t carry any priceless data. However, the methods to soundly scale back the variety of options in a ML downside aren’t trivial and want a proof that I’ll present on this submit.Image by the writer.The instruments I’ll current not solely simplify the computation effort and enhance the prediction accuracy, however they can even function a device to graphically visualize high-dimensional knowledge. For…
https://towardsdatascience.com/dimensionality-reduction-made-simple-pca-theory-and-scikit-learn-implementation-9d07a388df9e
Dimensionality Reduction Made Simple: PCA Theory and Scikit-Learn Implementation | by Riccardo Andreoni | Feb, 2024
