Chrysolites Tech Academy

Video by freeCodeCamp.org · streamed from YouTube. All rights remain with the creator.

Lesson 4: Machine Learning with Scikit-learn

About this lesson

This course, created by DLAcademy, teaches machine learning in Python with scikit-learn, starting from installation and building up to neural networks and image recognition. It opens by installing the library and its companions and plotting a first graph, then explains how a dataset is divided into features, the inputs a model reads, and labels, the answers it learns to predict. Early on you learn how to save a trained model to a file and open it again, so that training does not have to be repeated every time a program runs. The course then introduces classification and the train and test split, the practice of holding back part of the data to measure a model fairly. A series of algorithms follows, each explained and then coded with the library: k-nearest neighbours, support vector machines, linear regression, a comparison of logistic and linear regression that clarifies when each one applies, and k-means clustering for finding groups in unlabelled data. The second part turns to neural networks. It explains how they are built, what overfitting and underfitting look like and how to recognise them, and how networks learn through backpropagation, a cost function and gradient descent. Convolutional neural networks are introduced next as the architecture designed for images. The course ends by implementing a recogniser for handwritten digits, bringing together the whole workflow. Afterwards, read Chapter 4 of the course book, which works a convolution by hand.

Install Chrysolites Tech Academy

Android / Chrome / Edge: open the browser menu and choose Install app or Add to Home screen.

iPhone / iPad (Safari): tap the Share button, then Add to Home Screen.