Spaces:
Runtime error
Runtime error
| # Copyright 2019 Seth V. Neel, Michael J. Kearns, Aaron L. Roth, Zhiwei Steven Wu | |
| # | |
| # Licensed under the Apache License, Version 2.0 (the "License"); you may not | |
| # use this file except in compliance with the License. You may obtain a copy of | |
| # the License at http://www.apache.org/licenses/LICENSE-2.0 | |
| # | |
| # Unless required by applicable law or agreed to in writing, software distributed | |
| # under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR | |
| # CONDITIONS OF ANY KIND, either express or implied. See the License for the | |
| # specific language governing permissions and limitations under the License. | |
| """Function generating 3-d heatmap visualizing gamma-disparity. | |
| The main function in this module, heat_map(), generates and saves a 3-d | |
| heatmap visualizing the gamma-disparity for groups defined by linear thresholds over 2 sensitive attributes. | |
| This serves as a (heuristic) method to help visualize convergence of the algorithm via brute force checking in | |
| low dimensions, rather than relying on the Auditor. See [KRNW18] for details. | |
| """ | |
| import matplotlib | |
| import pandas as pd | |
| import matplotlib.pyplot as plt | |
| from matplotlib import cm | |
| from mpl_toolkits.mplot3d import Axes3D | |
| from aif360.algorithms.inprocessing.gerryfair.reg_oracle_class import * | |
| def calc_disp(predictions, X, group_labels, X_prime, group): | |
| """Return the fp disparity in a group g - helper function for heat_map. | |
| Args: | |
| :param predictions: dataframe of predictions of the classifier | |
| :param X: dataframe of covariates | |
| :param group_labels: dataframe of group labels | |
| :param X_prime: sensitive covariates | |
| :param group: object of class Group(), see auditor.py | |
| Returns: | |
| :return: weighted disparity on the group g | |
| """ | |
| X_0 = pd.DataFrame( | |
| [X_prime.iloc[u, :] for u, s in enumerate(group_labels) if s == 0]) | |
| group_0 = group.predict(X_0) | |
| n = len(group_labels) | |
| g_size_0 = np.sum(group_0) * 1.0 / n | |
| FP = [predictions[i] for i, c in enumerate(group_labels) if c == 0] | |
| FP = np.mean(FP) | |
| group_members = group.predict(X_prime) | |
| fp_g = [ | |
| predictions[i] for i, c in enumerate(group_labels) | |
| if group_members[i] == 1 and c == 0 | |
| ] | |
| if len(fp_g) == 0: | |
| return 0 | |
| fp_g = np.mean(fp_g) | |
| return (FP - fp_g) * g_size_0 | |
| def heat_map(X, X_prime, y, predictions, eta, plot_path, vmin=None, vmax=None): | |
| """Generate 3-d heatmap and save it at plot_path. | |
| Args: | |
| :param eta: discretization parameter of coefficients defining subgroups | |
| :param plot_path: the path to save the heatmap at | |
| :param vmin: Min value to map: see plot_surface documentation in matplotlib | |
| :param vmax: Max value to map | |
| Returns: | |
| :return: the min and max gamma disparities on groups in the plot | |
| """ | |
| fig = plt.figure() | |
| ax = fig.add_subplot(1, 1, 1, projection='3d') | |
| columns = [str(c) for c in X_prime.columns] | |
| attribute_1 = np.zeros(int(1 / eta)) | |
| attribute_2 = np.zeros(int(1 / eta)) | |
| disparity = np.zeros((int(1 / eta), int(1 / eta))) | |
| for i in range(int(1 / eta)): | |
| for j in range(int(1 / eta)): | |
| beta = [-1 + 2 * eta * i, -1 + 2 * eta * j] | |
| group = LinearThresh(beta) | |
| attribute_1[i] = beta[0] | |
| attribute_2[j] = beta[1] | |
| disparity[i, j] = calc_disp(predictions, X, y, X_prime, group) | |
| X_plot, Y_plot = np.meshgrid(attribute_1, attribute_2) | |
| ax.set_xlabel(columns[0] + ' coefficient') | |
| ax.set_ylabel(columns[1] + ' coefficient') | |
| ax.set_zlabel('gamma disparity') | |
| ax.set_zlim3d([np.min(disparity), np.max(disparity)]) | |
| surface = ax.plot_surface(X_plot, | |
| Y_plot, | |
| disparity, | |
| cmap=cm.coolwarm, | |
| linewidth=0, | |
| antialiased=False, | |
| vmin=vmin, | |
| vmax=vmax) | |
| if plot_path != '.': | |
| fig.savefig('{}.png'.format(plot_path)) | |
| plt.close() | |
| return [np.min(disparity), np.max(disparity)] | |