Do you know how to use one of the most popular machine learning libraries for Python? Master scikit-learn through a combination of lecture and hands-on (via Jupyter) in this eight-part video series:
- Scikit-learn Overview. This first video in the scikit-learn series explains the use cases for scikit-learn, and provides an overview to its regression, classification, and clustering algorithms. Most of scikit-learn is written in Python, but some (where high performance is needed) is written in Cython. Learn about the core of scikit-learn which is the estimator API. Learn why scikit-learn is easy to use and the four steps to follow in using scikit-learn.
- Installing Scikit-learn. This second video in the scikit-learn series outlines the scikit-learn installation steps. Scikit-learn requires Python, NumPy, and SciPy.
- Loading Data Sets using Scikit-learn. This third video in the scikit-learn series shows you the three ways to load data using scikit-learn. Learn about Sylearn, which accepts data as either a numpy array or a pandas data frame.
- Pre-processing Data using Scikit-learn. This fourth video in the scikit-learn series explains how to apply transformations to the data before feeding data to the algorithm. Learn how to use sklearn, and the six steps to follow in pre-processing data: mean removal and variance scaling, non-linear transformation, normalization, encoding categorical features, discretization, and the imputation of missing values.
- Splitting Data into Train Sets and Test Sets in Scikit-learn. This fifth video in the scikit-learn series shows you how to perform Train-Test-Split in scikit-learn. Learn why it is important to split your data in a random manner. Apply the various parameters such as test_size and train_size that are required to perform Train-Test-Split.
- Linear Regression using Scikit-learn. This sixth video in the scikit-learn series shows you how to apply linear regression in scikit-learn. Linear regression is a statistical model that is used for finding linear relationships between a target and one or more predictors. Learn how to set up dependent variables and independent variables and the two types of linear regression (simple linear regression and multiple linear regression). We explore the R-squared statistical measure and Root Mean Squared Error.
- Naïve Bayes using Scikit-learn. This seventh video in the scikit-learn series outlines how to apply the Naïve Bayes classifier in scikit-learn. Understand the concept of feature independence and Bayes Theorem of probability.
- SVM using Scikit-learn. This eighth video in the scikit-learn series shows you how to apply the Support Vector Machines supervised machine learning algorithm in scikit-learn. SVM is a non-probabilisitic classifier model. Learn about SVM (Support Vector Machine), SVR (Support Vector Regression), and SVC (Support Vector Clustering).