forked from sourenaKhanzadeh/face_detection_ml_matlab
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrecog_summary.m
21 lines (19 loc) · 1.26 KB
/
recog_summary.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
%% CPS843/CP8307 Assignment 3 - MATLAB Machine Learning
% Include this file in your submission. Do not modify it, other than to add
% your name and student number. The grader should be able to run this
% script to step through the entire assignment.
% - For multi-part questions, insert your own pauses, where appropriate.
% - For written answer questions, print your answer in the MATLAB terminal.
% Student Name: Sourena Khanzadeh
% Student Number: 500929191
% Student Name: Dylan Donsky
% Student Number: 500700489
fprintf("for our training set and creating of the svm we pretty much just \n")
fprintf("created four seperate image folders, two for training negative and positive images \n")
fprintf("and two for the validation set of negative and postive images. \n")
fprintf("for our cell size we shortened it from the default of 6 to 3\n")
fprintf("in order to get a sharper and more accurate pattern.\n")
fprintf("for our lambda values we went with a minimal balue of 0.00005 to make it more precise.\n")
fprintf("our best accuracy for our validation set is as follows\n")
fprintf("Classifier performance on test data: \n accuracy: 0.997 \n true positive rate: 0.497\n")
fprintf(" false positive rate: 0.000 \n true negative rate: 0.500 \n false negative rate: 0.003\n")