In this post, I explain what PCA is, when, and why to use it, and how to implement it in Python using scikit-learn. Also, I explain how to get the feature importance after a PCA analysis. — 1. Introduction & Background Principal Components Analysis (PCA) is a well-known unsupervised dimensionality reduction technique that constructs relevant features/variables through linear (linear PCA) or non-linear (kernel PCA) combinations of the original variables (features). In this post, we will only focus on the famous and widely used linear PCA method.

In this article I explain the core of the SVMs, why and how to use them. Additionally, I show how to plot the support vectors and the decision boundaries in 2D and 3D. — Introduction Everyone has heard about the famous and widely-used Support Vector Machines (SVMs). The original SVM algorithm was invented by Vladimir N. Vapnik and Alexey Ya. Chervonenkis in 1963. SVMs are supervised machine learning models that are usually employed for classification (SVC — Support Vector Classification) or regression (SVR — Support…

Mathematical formulation, Finding the optimum number of clusters and a working example in Python — Introduction K-means is one of the most widely used unsupervised clustering methods. The K-means algorithm clusters the data at hand by trying to separate samples into K groups of equal variance, minimizing a criterion known as the inertia or within-cluster sum-of-squares. This algorithm requires the number of clusters to be specified…

What is a Multilayer Perceptron? What are the pros and cons of MLP? Can we classify handwritten digits accurately using a MLP classifier? How do learnt weights look like? — 1. Short Introduction 1.1 What is a Multilayer Perceptron (MLP)? An MLP is a supervised machine learning (ML) algorithm that belongs in the class of feedforward artificial neural networks [1]. The algorithm essentially is trained on the data in order to learn a function. Given a set of features and a target variable (e.g. labels) it learns a non-linear function…

Unlock the Power of Data Science: From Custom Model Fitting to Career Advice, We’ve Got You Covered — Introduction After years of passionate engagement behind the scenes, we’re thrilled to announce the launch of our brand new initiative, the “Data Science Hub” on Patreon (https://www.patreon.com/TheDataScienceHub). …

Using the ChatGPT that everyone now knows about it via the terminal! — 1. Introduction In recent years, OpenAI’s ChatGPT has taken the world of Artificial Intelligence by storm. As a highly advanced language model, ChatGPT has the ability to generate human-like text and perform various NLP tasks with remarkable accuracy. In this article, we will explore 5 of the most impressive things ChatGPT can…

In this post, I show you how to predict stock prices using a forecasting LSTM model & a simple Ridge regression model. — 1. Introduction 1.1. Time-series & forecasting models Most machine learning models use observations without a time dimension. Time-series forecasting models predict future values based on previously observed values and are useful for non-stationary data. Non-stationary data, whose statistical properties vary over time, are commonly referred to as time-series, such as temperature, stock prices, and house prices over…

Here is my personal view based on recent experience after getting interviewed by 4 companies and finally getting 2 offers for a Data Scientist role. Introduction Data Science is a fast-growing field, and with the increasing demand for data scientists, it’s important to know how to ace the job interview process…

Here is my personal view based on my recent experience after getting interviewed by 4 companies and finally getting 2 offers for a Data Scientist role. Introduction Data Science is a fast-growing field, and with the increasing demand for data scientists, it’s important to know how to ace the job interview…