> For the complete documentation index, see [llms.txt](https://kflagg.gitbook.io/pythonds/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://kflagg.gitbook.io/pythonds/2.-pandas-structuring-data-for-analysis.md).

# 2. pandas: Structuring Data For Analysis

## Objectives

* Create `pandas` data frames.
* Sort and access data stored in data frames.

`pandas` is a Python module that provides functionality for data organized in tabular form (like a spreadsheet). The tabular data layout, called a data frame, is the bread and butter of data science, and likewise `pandas` is central to working with tabular data in Python.

If you are eager to see what `pandas` can do, have a look at the Pandas Cookbook for a crash course. For a gentler but deeper experience, skip to the selections from the official `pandas` documentation.

## Optional Crash Course: Pandas Cookbook

If you are ready to dive in, the [Pandas Cookbook](https://github.com/jvns/pandas-cookbook) provides a broad overview of common tasks in `pandas`. The cookbook includes data files and a collection of IPython notebooks organized into chapters. Just open each notebook in Jupyter and work through Chapters 1-4 and 6-8 in order. Expect each chapter to take 5 to 20 minutes.

The cookbook was written for Python 2 and older versions of`pandas` so I found a few things that did not quite work on my system. Try debugging it as practice interpreting Python error messages!

The two places I encountered big problems were in Chapters 5 and 9. Chapter 5 relies on a web data source that has either moved or changed its interface so that the code no longer works. You can read through the text in Chapter 5 if it sounds interesting. Chapter 9 is an example of connecting to databases, and is incomplete. You don't need that skill yet and there are good resources elsewhere. Skip it.

## Deeper Intro: 10 minutes to pandas

The `pandas` documentation also includes a quick-start called [10 minutes to `pandas`](https://pandas.pydata.org/docs/getting_started/10min.html). It has grown considerably longer than ten minutes (it took me about 4 hours to get through). It is a rather technical gallery of concise examples showing how do many simple tasks.

## Selections From The `pandas` Documentation

The full `pandas` documentation gives detailed example walkthroughs. I recommend the following progression.

1. `pandas` Data Structures: [Intro to data structures](https://pandas.pydata.org/docs/getting_started/dsintro.html)
2. Essentials: [Essential basic functionality](https://pandas.pydata.org/docs/getting_started/basics.html)
3. Reading And Writing Data: [How do I read and write tabular data?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/02_read_write.html)
4. Subsetting: [How do I select a subset of a `DataFrame`?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/03_subset_data.html)
5. Deriving Features: [How to create new columns derived from existing columns?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/05_add_columns.html)
6. Rearranging a `DataFrame`: [How to reshape the layout of tables?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/07_reshape_table_layout.html)
7. Combining more than one `DataFrame` [How to combine data from multiple tables?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/08_combine_dataframes.html)
8. Aggregating And Summarizing [How to calculate summary statistics?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/06_calculate_statistics.html)
9. Dates And Times: [How to handle time series data with ease?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/09_timeseries.html)
10. Text: [How to manipulate textual data?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/10_text_data.html)
11. Plots: [How to create plots in `pandas`?](https://pandas.pydata.org/docs/getting_started/intro_tutorials/04_plotting.html)