Created
April 25, 2019 15:05
-
-
Save njvack/8b1c6d1ffc829bde337f6610db5ccf4a to your computer and use it in GitHub Desktop.
Match distributions of two pandas dataframes
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # -*- coding: utf-8 -*- | |
| # Written by Nate Vack <njvack@wisc.edu> at the Center for Healthy Minds | |
| # Copyright 2019 Board of Regents of the University of Wisconsin System | |
| # Permission is hereby granted, free of charge, to any person obtaining a copy | |
| # of this software and associated documentation files (the "Software"), to deal | |
| # in the Software without restriction, including without limitation the rights | |
| # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell | |
| # copies of the Software, and to permit persons to whom the Software is | |
| # furnished to do so, subject to the following conditions: | |
| # The above copyright notice and this permission notice shall be included in | |
| # all copies or substantial portions of the Software. | |
| # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR | |
| # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, | |
| # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE | |
| # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER | |
| # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, | |
| # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN | |
| # THE SOFTWARE. | |
| import numpy as np | |
| def distance(point1, point2): | |
| return np.sqrt(np.sum((point1 - point2) ** 2)) | |
| def euclidian_distribution_match( | |
| distribution_data, sample_data, fields, debug=False): | |
| """ | |
| Given distribution_data, sample_data, and a list of fields, return a new | |
| dataframe, sampled from sample_data, of the same size as distribution_data. | |
| Each point in the returned dataset will be the closest match, by | |
| euclidean distance, in sample_data. sample_data is sampled without | |
| replacement. | |
| This algorithm is super slow, you could doubtless make it faster. | |
| """ | |
| samp = sample_data.copy() | |
| samp['_match_index'] = None | |
| for index, row in distribution_data.iterrows(): | |
| point1 = row[fields] | |
| def dist_fx(point2): | |
| return distance(point1.values, point2.values) | |
| samp['_difference'] = samp[fields].apply(dist_fx, axis=1) | |
| samp.sort_values(by='_difference', inplace=True) | |
| match = samp.loc[samp['_match_index'].isnull()].iloc[0] | |
| match_index = match.name | |
| if debug: | |
| print(f'Search {point1}, found {match[fields]} at {match_index}') | |
| samp.loc[match_index, '_match_index'] = index | |
| return samp.loc[~samp['_match_index'].isnull()] | |
| def match_distributions(distribution_data, sample_data, field_name): | |
| """ | |
| Given distribution_data, sample_data, and a field_name, return a new | |
| dataframe, sampled from sample_data, of the same size as distribution_data | |
| and as closely matched on the distribution of `field_name` as possible. | |
| This is a (much faster) special-case of the euclidian match above. | |
| It's also an inefficient algorithm, though! | |
| """ | |
| samp = sample_data.copy() | |
| samp['_match_index'] = None | |
| for index, row in distribution_data.iterrows(): | |
| val = row[field_name] | |
| samp['_difference'] = (samp[field_name] - val).abs() | |
| samp.sort_values(by=['_difference'], inplace=True) | |
| match = samp.loc[samp['_match_index'].isnull()].iloc[0] | |
| match_index = match.name | |
| samp.loc[match_index, '_match_index'] = index | |
| return samp.loc[samp['_match_index'] > 0] |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment