Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Subscribe
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Practical Machine Learning with Scikit-Learn
Rating: 4.1 out of 5(364 ratings)
16,187 students

Practical Machine Learning with Scikit-Learn

Learn the most powerful machine learning algorithms in under an hour
Created byAdam Eubanks
Last updated 6/2020
English

What you'll learn

  • How to implement regression, classification and boosting algorithms
  • Which algorithms work best for a given dataset
  • Data preprocessing

Course content

4 sections • 5 lectures • 1h 7m total length

  • Introduction1:59

    Explore practical machine learning with scikit-learn, covering data preprocessing, missing values, regression, classification, boosting, and principal component analysis using Google Colab and Python.

  • Data Preprocessing7:17

    Master data preprocessing by handling missing data with mean or median imputation, applying one-hot encoding for categorical variables, and scaling features to prevent overfitting.

Requirements

  • Basic python knowledge
  • Google Colab account

Description

Machine learning is a rapidly growing field. However, a lot of courses on the internet today do not go over some of it's most powerful algorithms. In this course, we will learn multiple machine learning algorithms, along with data preprocessing, all in under an hour. We will go over regression, classification, component analysis and boosting all in scikit-learn, one of the most popular machine learning libraries for python.

Algorithms we'll go over (in order):

  • Linear Regression

  • Polynomial Regression

  • Multiple Linear Regression

  • Logistic Regression

  • Support Vector Machines

  • Decision Trees

  • Random Forest

  • Principle Component Analysis

  • Gradient Boosting

  • XGBoost

Who this course is for:

  • People looking to get into AI but don't know where to start
  • People who want to build accurate models as quickly as possible

Report abuse