Principal Component Analysis in Python

In this video we take a look at Principal Component Analysis (PCA) and some of the math behind it including Eigenvalues, Eigenvectors, Singular Values and computing variation. After understand this and going through an example with 2 principal components, we move to using PCA in Python by using the sklearn implementation with the iris flower dataset. Then we practice visualizing the results with plotly. We also discuss the importance of normalizing your dataset for PCA and in general for machine learning and I show you how to do so via standardization and min-max normalization. Finally we revisit the crypto dataset from previous videos and apply PCA for dimensionality reduction and look at the results.

My initial example is heavily based off of StatQuest's great video on PCA. Check it out:
StatQuest. “Principal Component Analysis (PCA), Step-by-Step” YouTube, Joshua Starmer, 2 Apr. 2018, https://youtu.be/FgakZw6K1QQ

Useful Reference(s):
Scikit-Learn Documentation on the Principal Component Analysis: https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html
Scikit-Learn PCA Example on the Iris Dataset: https://scikit-learn.org/stable/auto_examples/decomposition/plot_pca_iris.htm #sphx-glr-auto-examples-decomposition-plot-pca-iris-py
Plotly Graphing Library: PCA Visualization in Python: https://plotly.com/python/pca-visualization/
StatQuest: PCA in R: https://youtu.be/0Jp4gsfOLMs
A good reference on PCA with examples. Also source of the gif rotating image: https://builtin.com/data-science/step-step-explanation-principal-component-analysis
More tutorials of PCA: https://towardsdatascience.com/pca-using-python-scikit-learn-e653f8989e60
A nice article of PCA and finance: https://riskprep.com/tutorials/understanding-principal-component-analysis/
PCA vs t-SNE (if you want to learn other algorithms of the same type: https://medium.com/analytics-vidhya/pca-vs-t-sne-17bcd882bf3d
Scikit-Learn Documentation on Normalization: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.normalize.html
Tutorial on how to Normalize data in Python: https://www.journaldev.com/45109/normalize-data-in-python
Standardization Reference: https://www.statisticshowto.com/standardized-values-examples/
The Crypto Dataset: https://www.kaggle.com/odins0n/top-50-cryptocurrency-historical-prices
Previous Videos: https://www.youtube.com/watch?v=tPkJ1oDNW3U&list=PLrosN5xeGOCtc998ALNlMMPMti54P0bGk&index=12



Credits where Credits are Due:
-Subscribe End Screen: Graphicki https://www.youtube.com/watch?v=zv8Ek1dBJTc
-Live Animation: Motion Chroma | Leader https://www.youtube.com/watch?v=80wavzrN00c
-Music: Song: Without You, Artist: MNRVA, youtube.com/XimerTracks
-Streaming Software: OBS


Thank you for watching!