Package management

Each notebook is backed by a python kernel, or execution backend. All notebook kernels share the same Python virtual environment . These means that when a package such as pandas is installed in a notebook, it is available to all other notebooks in your project. We recommend using a dedicated notebook for installing and managing dependencies.

To install a package in your project, execute a cell containing:

!uv pip install <package>

or:

!pip install <package>

DataCards uses the uv package manager under the hood, so whichever command you use, the effect will be the same. Packages are installed globally and are available to all notebooks in the project.

httpx
jinja
matplotlib
numpy
polars
pandas
scikit-learn
pillow