This blog post is intended for those who do not have a background in programming, data and/or statistics.

Machine Learning is a complex subject that will never be covered in just one blog post. And it would be difficult to cover, learn and fully comprehend it within 3 days.

Yet that’s what we did.

I struggled to get my model off the ground to begin with so I am sharing 3 tips on how to approach the “Machine Learning” week at The Data School to have an easier time.

1. Don’t be afraid to ask for help and ask questions early on

This advice goes for any subject, but it can apply here especially if you are new to the data industry. There is a lot of information to grasp and wrap your head around.

2. When exploring your data relate it to your Target (aka the “y” or outcome that we are striving to predict)

For your EDA (Exploratory Data Analysis), you will attempting to find connections, correlations, distributions in order to fill in missing values or determine the best “features” (aka categories, columns or fields) to plug into the model.

Alteryx and Tableau both have tools that can assist you with this.

In Alteryx you there is a “Data Investigation” tab that gives you access to a variety of analysis tools including Histogram and Field Summary.

Helpful for EDAs


And in Tableau, the “Show Me” tab will also provide you with histograms and box plots. Just make sure you can relate it to the “Target” variable.

3. The formatting of your Training data needs to be the same as the Testing data.

This was what prevented me from creating a working model.

One method of overcoming this is to combine the train and test data together then split it out. The Test data target variable will have “NULL” or missing values so after you’re done with feature engineering, you can filter the Test data out that way.

This is just an example.

BONUS TIP! Get your model working and then go back to tweak it!

Rather than try to have EVERY feature lined up and putting it through a model, put through a few features first. That way you can always tweak it whenever there are errors.


I hope you found these tips were useful.

If you would like to contact me, please feel free to connect with and message me over at LinkedIn.

Thank you for reading and have a great day!

The Data School
Author: The Data School