Gain insights into data science with easy-to-follow, hands-on explanations. Explore essential concepts quickly and efficiently, even without prior statistics knowledge, for a career boost.

GrokkingDataScientist.tar.gz

jupyter_job

Master the skills that can get you a $100K+ salary even if you bunked your statistics classes. 

No need to waste hours and hours on browsing from one article to the next and piecing together the info you need to grasp important topics. No need to get overwhelmed by the information overload. Find easy to follow, hands-on, and fun explanations of all the essential topics in one place so you can quickly and efficiently learn what you need to thrive as a data scientist.

"Is this course right for me?" Continue to read to decide for yourself!

- "I want to understand this data science concept. Let me Google it". Then after hours of surfing, reading random articles, and invoking the heavens, you are more confused than before.
- "Data science is the sexiest and highest paying job of the 21st century. I want to become a data scientist too".
- "I have a basic knowledge of Python, willingness to learn, and commitment to become a great data scientist."

Is that you? If yes, you are at the right place.

Grokking Data Science

# 6. Grouping

Things start looking really interesting when we group rows with certain criteria and then aggregate their data. 

Say we want to group our dataset by director and see **how much revenue (sum) each director earned at the box-office and then also look at the average rating (mean) for each director**. We can do this by using the`groupby` operation on the column of interest, followed by the appropriate aggregate (*sum/mean*), like so:


# Let’s group our dataset by director and see how much revenue each director has
movies_df.groupby('Director').sum()

# Let’s group our dataset by director and see the average rating of each director
movies_df.groupby('Director')[['Rating']].mean()

As we can see, Pandas grouped all the <i>'Director'</i> rows by name into one. And since we used *sum()* for aggregation, it added together all the numerical columns. The values for each of the columns now represent the sum of values in that column for that director. 

For example, we can see that the director Aamir Khan has a very high average rating (8.5) but his revenue is much lower compared to many other directors (only 1.20M $). This can be attributed to the fact that we are looking at a dataset with international movies, and Hollywood directors/movies have understandably much higher revenues compared to movies from international directors.

In addition to `sum()` and `mean()` Pandas provides multiple other aggregation functions like `min()`and `max()`.

***Alt! Can you find a problem in the code when we apply aggregation to get the sum?***

This is not the correct approach for all the columns. We do not want to sum all the <i>'Year'</i> values, for instance. To make Pandas apply the aggregation on some of the columns only, we can specify the name of the columns we are interested in. For example, in the second example, we specifically passed the *'Rating'* column, so that the mean did not get applied to all the columns.

Grouping is an easy and extremely powerful data analysis method. We can use it to fold datasets and uncover insights from them, while aggregation  is one of the foundational tools of statistics. In fact, ***learning to use `groupby()` to its full potential can be one of the greatest uses of the Pandas library***.


Python Fundamentals for Data Science

The Fundamentals of Statistics

Machine Learning 101

End-to-End Machine Learning Project

The Real Talk

Pandas DataFrame Operations - Grouping and Sorting

6. Grouping