Machine Learning Journey

In this lecture you'll learn exactly what Simple Linear Regression models are used for.  We'll go through some practical use cases.  Then we'll get into the math behind the models, explain what the matrix of features X is and the dependent variable vector y.  You'll also learn how the models build the regression lines and how to interpret the output.  Let's do this baby!

Okay, now it's time to grab our dataset.  I'll introduce you to a popular online community of new and advanced data scientists which includes example dataset we can use to boost our skills.  Then we'll import our ML libraries and get ready for fun!

Okay, now you're going to use Pandas to import our Dataframe and split it into X and y.  We'll even start plotting our data points to make sure the dataset is suitable to Simple Linear Regression!  Let's go!

X_train? X_test? y_train? y_test? What the heck am I talking about? Come inside to see!

Now it's time to actually train our ML Models.  In this lecture, you will train your very first model!  It will be an exhilarating experience! haha.  Not only will you train your first model, you'll also learn how to prove accuracy and have complete confidence that your model actually works!  What could be better than that!? Let's go!

Now it's time to visualize your hard work.  We're going to use the Python Matplotlib module to build beautiful visualizations of our data.  You'll see both the real data and the predicted data so you can easily understand the relationships between the data elements so you can provide real value to stakeholders.

We're going to round out this section with an intuitive understanding of the math behind our Simple Linear Regression model.  This is a fun one because it will tie together everything you've been doing and really help you make sense of how all the pieces fit together!

Now it's time to learn about Multiple Linear Regression, how it differs from Simple Linear Regression and when you would use it.  We'll also reviewing the three main categories of machine learning algorithms and even discuss some pitfalls to watch out for. 

This lecture is going to be awesome.  We are going to role play a realistic use case for Multiple Linear Regression.  By the end of this lecture you'll not only grasp the math behind Multiple Linear Regression but also get a firm understanding on the true power of this amazing prediction algorithm!

In this lightning fast lecture we'll create our Jupyter notebook, import our libraries and execute our notebook in preparation for the data import stage!  Let's gooooo! lol.

Now it's time to bring in our dataset.  In this lecture we have also have a special surprise: you're going to learn how to use the Seaborn module to get instant analysis of your dataset via a beautiful matrix of colors and numbers!  I can't wait to show you how to do this!  Let's jump in - NOW!

Okay now it's time to encoding our categorical data.  You're not only going to learn how to do it, I'm also going to teach you how to use the integrated help to understand the syntax and function templates so you can because a true Machine Learning wizard.  I'm also going to teach you the differences between Python Lists and Tuples because you'll need to understand this to use our category encoding code.

Test sizes, independent variables, dependent variables, random seeds, what's going on here?? You'll know how to horizontally split and scientifically test your datasets after this lesson! Are you ready??

Now it's time to build and teach our multiple linear regression model.  And the best part you're going to understand the programmatic difference between the two.. in less than 4 minutes! YES!! Let's go!

Predicting the future? Wha??? Okay, so now it's time to get what you came for.  All prior lectures led up to this one: predicting new data outcomes.  You're going to learn how to use the predict method, the reshape method, the concatenate method, the R squared method, the .... oh man there's so much good stuff here! Let me just quit talking and start showing.  Let's go!

In this lesson you'll see how Polynomial Regression differs from Multiple Linear Regression and Simple Linear Regression. We're going to spend the entire lecture understanding the math and graphs in a way that is both intuitive and easy to assimilate.  We're in the deep end of the pool now and there's no holding back!  It's time to swim baby!

Let's look at our problem set: We're going to set the stage for our real world use case for Polynomial Regression by analyzing the dataset and making sense of our goals.

You already know what to do: importing Python libraries is breeze now.  Let's knock it out in 3 minutes.

Today you will no only learn how to import the dataset but also how to quickly inspect the data with the info and head methods! 

Now it's time to supercharge our dumb models into smart models.  But that's not all: in this lecture we're going to build two models: one simple linear regression model and one polynomial regression model.  Why? So you can see how each performs against our dataset and why our Polynomial Regression Model is best adapted to make the most accurate predictions!

Now it's time to plot our data.  As a bonus, I'm going to show you how to bump up the curve resolution for smoother lines and cleaner predictions. 

Had to save the best for last.  I'm going to patiently walk you through using our Polynomial Regression Model.  And I'm going to do it in a way that anticipates common issues you will experience.  I'm going to get in front of those pitfalls and show you how to think through the red errors and quickly cook up clean code!  Let's do it baby!