4.5 out of 5
4.5
71 reviews on Udemy

End-to-end Machine Learning: Decision trees

Build a transit time predictor in python
Instructor:
Brandon Rohrer
772 students enrolled
English [Auto-generated]
Build a decision tree from scratch.
Tune a machine learning model's hyperparameters and evaluate its performance.
Use Pandas DataFrames to manipulate data and make statistical computations.
Create a custom model visualization using matplotlib.
Use decision trees to make predictions.
Build your own transit time predictor.
Handle dates and times cleanly using python's datetime package.
Save and restore data using python's pickle package.
Programmatically dowload and parse data in a JSON format.
Interpret command line arguments using python's argparse package.

In this course, we’ll build and use decision trees, a popular and versatile tool that will serve you well in your applied machine learning work.

The data science problem we want to solve is predicting transit times on a public transportation system. We will walk through the entire process from end to end:

  1. Define the problem

  2. Gather the data

  3. Clean and prepare the data

  4. Build a decision tree

  5. Train the model

  6. Generate predictions

  7. Wrap the model in some code that makes it easy to use

If you are a professor or a teacher at any level, you are welcome to evaluate the course for free, and I can set your students up with a deep educational discount. Just contact me for the coupon code (brohrer@gmail.com).

I hope you enjoy the process of building a complete solution to a data science problem from the ground up.

Get the data

1
Introduction
2
Define the problem
3
Find the data
4
Handle datetimes
5
Gather the data
6
Timezone bug fix (addendum)

Prepare the data for modeling

1
Transform the data
2
Indexing pandas DataFrames
3
Archive the data
4
Visualize the data
5
Calculate possible arrival times
6
Create features

Build a model

1
How decision trees work
2
Build a decision tree
3
Build the tree nodes
4
Evaluate potential node splits
5
Visualize the tree
6
Inspect the tree

Put the model to work

1
Test the model
2
Tune the model
3
Get recommendations
4
Wrap up
You can view and review the lecture materials indefinitely, like an on-demand channel.
Definitely! If you have an internet connection, courses on Udemy are available on any device at any time. If you don't have an internet connection, some instructors also let their students download course lectures. That's up to the instructor though, so make sure you get on their good side!
4.5
4.5 out of 5
71 Ratings

Detailed Rating

Stars 5
37
Stars 4
25
Stars 3
7
Stars 2
1
Stars 1
1
face076a3840826e31a1f7127a01a5ec
30-Day Money-Back Guarantee

Includes

3 hours on-demand video
Full lifetime access
Access on mobile and TV
Certificate of Completion