Baby steps in learning Python for data analysis

Last weekend turned out to be a very special one!

My 10 month old daughter took her first baby steps and watching her take those steps was one of the most beautiful moment of my life. A baby, brimming with excitement to reach out to her father, trying to balance, while exploring her newly acquired skill and trying to speak simultaneously in her own way – was a moment to cherish! I hope I could have recorded that instance!

What followed the first little walk was equally exciting – Jenika (my daughter) was enjoying practicing her new skill in spite of all the limitations, she faced. She kept faltering every 3-4 steps, but it didn’t seem to affect her. There were a few things, which stood out in the manner, she was practicing her new skill:

1. She was enjoying the process.
2. The outcome of previous attempt didn’t matter. Every attempt was a fresh start.
3. Most of the time her focus was on the immediate next step rather than the end of the journey.

After spending the weekend with Jenika, both of us are still not over with the fun! She is still enjoying her day in the same innocent manner and I can’t stop thinking about it.

This is when this thought came to my mind – why not create a series of articles like baby steps – articles which enable users to take small steps towards a new skill, keep the excitement up and provide a refreshing start. I knew I was doing it, the second this thought crossed my mind!

The next question was, which area should these articles be written on? There were 2 areas I considered – Python for data analysis and Big data. Why them? For a simple reason, we have not spent a lot of time learning these skills till now. I will start with Python and we will take Big data at a later date.

Baby steps with Python for data analysis

In this series of articles, we will take bite sized information about how to use Python for data analysis, chew it till we are comfortable and practice it at our own end.

In today’s article, we will cover:

• Why learn Python for data analysis?
• How to install Python?
• Running a few simple programs in Python.
• Quick and dirty build of first model – just to illustrate, how easy it is to build a model in Python.

Why learn Python for data analysis?

Python has gathered a lot of interest recently as a choice of language for data analysis. I had compared it against SAS & R some time back. Here are some reasons which go in favour of learning Python:

• Open Source – free to install
• Awesome online community
• Very easy to learn
• Can become a common language for data science and production of web based analytics products.

Needless to say, it still has a few drawbacks:

• It is an interpreted language rather than compiled language – hence might take up more CPU time. However, given the savings in programmer time (due to ease of learning), it might still be a good choice.

How to install Python?

There are 2 approaches to install Python:

• You can download Python directly from its project site and install individual components and libraries you want
• Alternately, you can download and install a package, which comes with pre-installed libraries. I would recommend Enthought Canopy Express to go ahead. Another option could be Anaconda.

Second method provides a hassle free installation and hence I’ll recommend that to beginners. The imitation of this approach is you have to wait for the entire package to be upgraded, even if you are interested in the latest version of a single library. It should not matter until and unless, until and unless, you are doing cutting edge statistical research.

Choosing a development environment:

Once you have installed Python, there are various options for choosing an environment. Here are the 3 most common options:

• Terminal / Shell based
• IDLE (default environment)
• iPython notebook – similar to markdown in R

IDLE editor for Python

While the right environment depends on your need, I personally prefer iPython Notebooks a lot. It provides a lot of good features for documenting while writing the code itself and you can choose to run the code in blocks (rather than the line by line execution)

We will use iPython environment for our future articles.

Warming up: Running a few programs in Python:

You can use Python as a simple calculator to start with:

A few things to note:

• You can start iPython notebook by writing “ipython notebook” on your terminal / cmd, depending on the OS you are working on
• You can name a iPython notebook by simply clicking on the name – Untitled0 in the above screenshot
• The interface shows In [*] for inputs and Out[*] for output.
• You can execute a code by pressing “Shift + Enter” or “ALT + Enter”, if you want to insert an additional row after.

A few additional things you can try are:

• Printing a message
• Creating and running loops

If you feel, it is overwhelming, don’t worry, we will do these step by step in coming days.

Building a simple Logistic Regression model in Python:

Following are the steps required to create the first Logistic Regression model:

• Import the required libraries
• Read the data
• Explore and clean the data
• Build Logistic Regression model
• Interpret the results

Since the purpose of this model is to just illustrate how to build a model in Python, I will take a clean dataset (remember dataset iris, which Tavish used in his previous article?) and just go ahead and build a model on the entire dataset. We will look at ways to split the data into test and train at a later point.

P.S. If you don’t understand all the steps today, don’t worry. I would suggest that you download and install iPython and just run this program as is to get used to the environment.

Here are the libraries and the dataset you will need:

Once you have read the dataset, you can print the dataset to explore it and build a Logistic Regression model on it:

And finally, you can compare the expected and predicted values:

That’s it. Your first model in Python is ready. You can now use model.predict(‘input data’) to make classifications basis the data provided. In the next article, we will show a way to share these notebooks with other people, accept contributions from other people and ways to version control it through github.

What do you think about using Python for data analysis? Did you find this tutorial and idea about this series useful? Do let me know, through comments below.

If you like what you just read & want to continue your analytics learning, subscribe to our emails, follow us on twitter or like our facebook page.

You might be interested in: Baby steps in Python: Part 2 (Libraries and data structure)

kunal