{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Running the webapp locally\n", "\n", "You can run the same [webapp available online](http://aequitas.dssg.io) locally.\n", "\n", "After installing Aequitas, run the following command in your terminal.\n", "\n", "```\n", "python -m serve\n", "```\n", "You should see th output:\n", "```\n", "* Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)\n", "```\n", "\n", "## Load data\n", "\n", "Use the `browse ...` button to select a properly formated csv. (see [input data](./input_data))\n", "Hit `upload`.\n", "\n", "## Customize report\n", "\n", "* choose attributes\n", " * these are automatically pulled out from the CSV\n", " \n", "* choose reference group\n", " * majority: for a given attribute use the majority group\n", " * min metric: for a given attribute use the group with the lowest scores as a baseline\n", " * custom group: pick groups to serve as reference \n", " \n", "* choose metrics you care about\n", " * Equal Parity\n", " * Proportional Parity\n", " * False Positive Parity\n", " * False Negative Parity\n", "\n", "* choose fairness threshold\n", "\n", "\n", " \n", "\n", "## Get 'The Bias Report'\n", "The webapp returns descriptive interpretation of the results along with three sets of tables.\n", "\n", "* Fairness Measures Results\n", "* Bias Metrics Results\n", "* Group Metrics Results\n", "\n", "More information about output [here](./output_data.html).\n" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }