Python: Working with Predictive Analytics (2019)

1h 22mAdvanced2019-08-12

Authors

Isil Berkun

Data Scientist at Intel Corp.

Course details

Data can tell many stories: where it came from and where it’s going. Predictive analytics gives programmers a tool to tell stories about the future: to extract usable information and make accurate predictions. These predictions, in turn, allow business to make more informed, impactful decisions. Join Isil Berkun, data scientist, to explore predictive analytics with Python. Discover how to prepare data—fill in missing values, perform feature scaling, and more—and use prebuilt Python libraries to make and evaluate prediction models. She describes what models to use when, and explains the concepts in such a way that you can immediately apply them to your own work. By the end of the course, you’ll be able to leverage Python libraries like pandas and NumPy and choose the right prediction models for your projects.

Learning objectives
Explain how predictive analytics can assist with decision-making.
Differentiate between the types of data that are used.
Apply the correct functions to Python code to produce optimal results.
Explain why data needs to be preprocessed before using predictive models.
Distinguish between the different predictive models available.

Skills covered

Data ModelingPythonProgramming LanguagesData ScienceOpen SourceSoftware DevelopmentDeep Dive (X:Y)

Concepts

0. Introduction

01 - Predict data in Python
02 - Road map

1. Data Preprocessing

03 - Differentiate data types
04 - Python libraries and data import
05 - Handling missing values
06 - Convert categorical data into numbers
07 - Divide the data into test and train
08 - Feature scaling

2. Prediction Models

09 - Introduction to predictive models
10 - Linear regression
11 - Polynomial regression
12 - Support Vector Regression (SVR)
13 - Decision tree regression
14 - Random forest regression
15 - Evaluation of predictive models
16 - Hyperparameter optimization
17 - Challenge - Hyperparameter optimization
18 - Solution - Hyperparameter optimization

Conclusion

19 - Next steps