scoreNeuralNet : calculates the scores of regions/ segments using a trained neural network Neural network was already trained using a trained Pattern Recognition Problem INPUT : x : input to the network, quantities regarding the segment or region. net : neural network with parameters optimized for different cells and imaging conditions OUTPUT : score : rawScore from -50 to 50. Above 0 is a good segment/region. Copyright (C) 2016 Wiggins Lab Written by Stella Stylianidou. University of Washington, 2016 This file is part of SuperSegger. SuperSegger is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. SuperSegger is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with SuperSegger. If not, see <http://www.gnu.org/licenses/>.
0001 function [rawScore] = scoreNeuralNet (x,net) 0002 % scoreNeuralNet : calculates the scores of regions/ segments using a trained neural network 0003 % Neural network was already trained using a trained Pattern Recognition Problem 0004 % 0005 % INPUT : 0006 % x : input to the network, quantities regarding the segment or region. 0007 % net : neural network with parameters optimized for different 0008 % cells and imaging conditions 0009 % 0010 % OUTPUT : 0011 % score : rawScore from -50 to 50. Above 0 is a good segment/region. 0012 % 0013 % Copyright (C) 2016 Wiggins Lab 0014 % Written by Stella Stylianidou. 0015 % University of Washington, 2016 0016 % This file is part of SuperSegger. 0017 % 0018 % SuperSegger is free software: you can redistribute it and/or modify 0019 % it under the terms of the GNU General Public License as published by 0020 % the Free Software Foundation, either version 3 of the License, or 0021 % (at your option) any later version. 0022 % 0023 % SuperSegger is distributed in the hope that it will be useful, 0024 % but WITHOUT ANY WARRANTY; without even the implied warranty of 0025 % MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 0026 % GNU General Public License for more details. 0027 % 0028 % You should have received a copy of the GNU General Public License 0029 % along with SuperSegger. If not, see <http://www.gnu.org/licenses/>. 0030 0031 x = x'; 0032 y = net(x); % same as y = net(x) but faster 0033 prob = y(2,:); % probabilities of second class (score = 1) 0034 0035 % because of the way scores were calculated in the past I will shift the 0036 % rawScore by .5 and multiply by 100 to make them spread out! 0037 rawScore = (prob - 0.5) * 100; 0038 0039 end 0040