iprofiler

Interactive code profiling widget for IPython + Jupyter Notebook.

Tested with Jupyter notebook server 4.1.0 and 4.2.0, Python 2.7 and 3.5, IPython 4.1 and IPython 4.2. May be compatible with earlier/later versions.

This software is a work in progress and is not yet stable/release ready.

Dependencies

• Cython
• Ipython/Jupyter
• Bokeh

Installation

For a development installation (requires npm):

$ git clone https://github.com/j-towns/iprofiler.git
$ cd iprofiler
$ pip install -e .
$ jupyter nbextension install --py --symlink --user iprofiler
$ jupyter nbextension enable --py --user iprofiler

Usage

Use

%load_ext iprofiler

to import the iprofiler to your Jupyter notebook, then use the iprofile line magic

%iprofile [statement]

to profile a statement, or the cell magic %%iprofile to profile a cell.