Whenever I am doing analysis with pandas my first goal is to get data into a panda’s DataFrame using one of the many available options. For the vast majority of instances, I use read_excel, read_csv, or read_sql.

However, there are instances when I just have a few lines of data or some calculations that I want to include in my analysis. In these cases it is helpful to know how to create DataFrames from standard python data structures such as lists or dictionaries. The basic process is not difficult but because there are several different options it is helpful to understand how each works. I can never remember whether I should use from_dict, from_records, from_items or the default DataFrame constructor. Normally, through some trial and error, I figure it out. Since it is still confusing to me, I thought I would walk through several examples below to clarify the different approaches. At the end of the article, I briefly show how this can be useful when generating Excel reports.

Read more...

Despite being over 1 year old, one of the most popular articles I have written is Overview of Python Visualization Tools. After these many months, it is one of my most frequently searched for, linked to and read article on this site. I think this fact speaks to hunger in the python community for one visualization tool to rise above the rest. I am not sure I want (or need) one to “win” but I do continue to watch the changes in this space with interest.

All of the tools I mentioned in the original article are still alive and many have changed quite a bit over the past year or so. Anyone looking for a visualization tool should investigate the options and see which ones meet their needs. They all have something to offer and different use-cases will drive different solutions.

In the spirit of keeping up with the latest options in this space, I recently heard about Altair which calls itself a “declarative statistical visualization library for Python.” One of the things that peaked my interest was that it is developed by Brian Granger and Jake Vanderplas. Brian is a core developer in the IPython project and very active in the scientific python community. Jake is also active in the scientific python community and has written a soon to be released O’Reilly book called Python Data Science Handbook. Both of these individuals are extremely accomplished and knowledgeable about python and the various tools in the python scientific ecosystem. Because of their backgrounds, I was very curious to see how they approached this problem.

Read more...

I recently had the pleasure of participating in a crowd-sourced data science competition in the Twin Cities called Analyze This! I wanted to share some of my thoughts and experiences on the process - especially how this challenge helped me learn more about how to apply data science theory and open source tools to real world problems.

I also hope this article can encourage others in the Twin Cities to participate in future events. For those of you not in the Minneapolis-St. Paul metro area, then maybe this can help motivate you to start up a similar event in your area. I thoroughly enjoyed the experience and got a lot out of the process. Read on for more details.

Read more...

I have heard from various people that my previous articles on common Excel tasks in pandas were useful in helping new pandas users translate Excel processes into equivalent pandas code. This article will continue that tradition by illustrating various pandas indexing examples using Excel’s Filter function as a model for understanding the process.

Read more...

I would like to offer this blog as platform for people to share their success stories with python. Over the past couple of weeks, I have had a handful of conversations related to the topic of how to get python implemented in an organization. In these conversations, I have noticed a lot of common themes related to getting the process started and sustaining it over time.

Read more...