Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Finding Actionable Insights using Keras Autoencoders
Rating: 4.6 out of 5(89 ratings)
3,785 students

Finding Actionable Insights using Keras Autoencoders

Using Autoencoders to Better Understand your Customers - Measuring Customer Credit Risk
Created byManuel Amunategui
Last updated 4/2020
English
English [Auto],

What you'll learn

  • Learn to build a Keras Autoencoder using Python
  • Learn to extract actionable insights from data using unsupervised and semi-supervised modeling
  • Learn to find anomalies in data

Course content

1 section • 7 lectures • 56m total length

  • Introduction1:08

    Find source code here:

    https://www.viralml.com/video-content.html?v=2UueN6lI62o

  • About me1:10
  • What is an Autoencoder and what is it good for?6:16
  • Preparing the Open Source Statlog - German Credit Data11:07
  • Quick classification look with AutoML5:28
  • Building our Keras Autoencoder19:14

    Build a Keras autoencoder on credit data, split 70/30, compress 61 features to 24 and 12, train with MSE, and use reconstruction error for anomaly-based credit risk insight.

  • Investigating anomalies11:52

    Attach reconstruction errors from a Keras autoencoder to the testing data using mean squared error to flag anomalies. Inspect top outliers against good-customer norms to inform business tailoring and retraining.

Requirements

  • You should know a little Python, how to install Python libraries, and how to use Jupyter notebook

Description

Please join me for another exciting data science class where we apply autoencoders or unsupervised learning towards the pursuit of knowledge.


Remember at the end of the day modeling and data science don't mean much if we can't extract actual insights to help guide our customers, our friends, the research community in the advancement of whatever it is they are after using data. Autoencoders can help you better understand your data, answer your questions, and even discover new ones! Please join me on this exciting adventure!

Who this course is for:

  • Anybody wanting to analyze data
  • Anybody wanting to perform anomaly detection
  • Anybody interested in Autoencoders and machine learning with Keras

Report abuse