Executioncount

Executioncount, usually referred to as execution_count in official documentation, is a field used by IPython and Jupyter notebooks to record the execution order of code cells. It is part of the notebook file format (.ipynb) and is associated with each code cell. The value is an integer representing the global execution step when the cell was last run, or null if the cell has not yet been executed.

In the notebook interface, a code cell displays a prompt such as In [n]:, where n corresponds to the cell’s last execution count. Corresponding outputs are labeled with Out [n], linking the result to the same execution event. The count increases with every execution of a code cell, reflecting the order in which code was executed across the notebook. Rerunning cells updates the relevant execution_count values to the new global step.

Semantically, execution_count is primarily a labeling mechanism rather than a computational feature. It helps users understand the sequence of results, reproduce experiments, and assess the relative freshness of outputs. The count is stored in the notebook’s JSON representation, alongside the cell’s source code and outputs. If a kernel is restarted, the execution_count values may no longer reflect the newest state until cells are re-executed.

Not all environments expose execution_count, and it is specific to code cells within IPython/Jupyter notebooks. It does not affect the execution of code itself; it serves as a bookkeeping aid for users reviewing and sharing interactive sessions. See also: IPython, Jupyter Notebook, notebook file format, In [n] and Out [n] prompts.