الصفوف والأعمدة والفهرس

A DataFrame is pandas' spreadsheet in code: rows and columns of data you can slice, sort, and compute on instantly. 📊

Every DataFrame has two axes: rows run top to bottom, columns run left to right. Together they form one tidy grid of values.

Each column holds one variable, like price or city, and shares a single data type. Think of a column as one measured thing.

Each row is one record: a customer, a day, a sale. Read a row across and you get the full story of one observation.

The index is the label down the left side that names every row. By default it counts 0, 1, 2, but it can be anything unique.

The column headers form their own label set, the columns attribute. You select data by these names instead of guessing positions.

The shape attribute returns rows and columns as a pair. shape[0] is your row count, shape[1] is your column count.

Pull a single column and pandas hands you a Series: a one-dimensional labeled array. A DataFrame is really a stack of Series.

You can promote a column to be the index when it identifies rows, like an order id. It makes lookups clearer and faster.

Pandas aligns and joins data by labels, not row order. That is why a thoughtful index quietly powers most of your analysis.

Need a plain counting index back? reset_index moves the current index into a column and gives you fresh 0-based labels.

Time to test the mental model of a DataFrame.

Rows are records, columns are variables, and the index labels every row. Know this map and the rest of pandas clicks. 🗺️