pandas
The most widely used Python library for working with tabular data: DataFrames, file readers and writers, grouping, joining and time series.
Not checked by a trusted editor yet Last edited Outdated?
pandas-dev/pandas
- Stars
- 50k
- Forks
- 20.5k
- Language
- Python
- Licence
- BSD-3-Clause
- Last push
- 15 Oct 2026
On GitHub: Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more
Open on GitHubpandas gives Python a DataFrame: a table with labeled columns, each holding one data type. It reads and writes CSV, Excel, JSON, Parquet, SQL databases and other formats. It can then filter, join, reshape, group and aggregate the data with short, chainable expressions.
Typical uses are cleaning data exports, preparing data for charts or machine learning, and replacing spreadsheet work with repeatable scripts. It sits on top of NumPy and works well with Jupyter notebooks, Matplotlib and scikit-learn. Install it with pip install pandas or through conda.
The project is developed in the open under the BSD 3-Clause license, with documentation at pandas.pydata.org. Large datasets that do not fit in memory are a weak spot. For those, people reach for chunked reading or for engines such as DuckDB or Polars, which can exchange data with pandas.
See also
- 0074 pandas basics: read a CSV, filter rows and summarize with groupby Load a CSV into a pandas DataFrame, inspect it, filter rows with conditions, and compute totals per group, with the SQL equivalent… Intermediate · How-to · Any system Article Data analysis · Files & automation 15 Oct 2026
History
Every saved edit is kept. The plot on the sleeve draws one line per edit. Open any edit to see exactly what changed.
- Commitpedia +853 chars