- Basic Python Knowledge Required
- 8 weeeks | 2 x 1.5hr classes per week
By the end of this course you will:
- build and evaluate basic supervised learning predictive models
- build and evaluate basic unsupervised learning predictive models
- learn effective recipes for loading data, handling text or numerical data, model selection, and dimensionality reduction.
Machine Learning is omnipresent. It is an integral part of companies, products, research projects, and applications of every size in every industry. It is also fun! With basic Python knowledge as a prerequisite, this course will teach you practical ways to build your own predictive models and machine learning solutions.
With all the data available today, machine learning applications are limited only by your imagination (Which is limitless)
|1||Why Machine Learning? Why Python? Build your first model||Get intro into what Machine Learning is about. Build a model for classifying Iris Species|
|2||Supervised Learning. Linear Models||Build two of the of most commonly used and successful models of machine learning - linear regression and logistic regression|
|3||Supervised Learning. Naive Bayes Classifier. Decision Trees||Take a more generalized view on linear models with Naive Bayes Classifier and build a model that learns a hierarchy of if/else questions, leading to a decision|
|4||Supervised Learning. Neural Networks||Build a neural network model to recognize hand-written digits|
|5||Unsupervised Learning. K-means and Agglomerative Clustering||Get intro into unsupervised learning and its problems. Build two most common clustering models|
|6||Data Representation and Feature Engineering||Data you get is not always perfect. It is never perfect. You will learn to make it more suitable for your future machine learning pursuits|
|7||Working with Text Data||Text is one of the most common types of data you can encounter. You will build a sentiment analysys model based on movies review.|
|8||Wrap-up. Show your skills||Choose a dataset. Come up with a cool problem. Solve it.|