{"id":61,"date":"2019-03-09T23:17:23","date_gmt":"2019-03-09T23:17:23","guid":{"rendered":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/?p=61"},"modified":"2019-03-09T23:17:23","modified_gmt":"2019-03-09T23:17:23","slug":"ryan-week-4-status-report","status":"publish","type":"post","link":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/2019\/03\/09\/ryan-week-4-status-report\/","title":{"rendered":"Ryan &#8211; Week 4 Status Report"},"content":{"rendered":"<p>For this week, I began classifying the heart sound training data using a CNN. I researched that CNN&#8217;s aren&#8217;t very optimal to use for audio files so instead I am taking the spectograms of the data and then training the network on the images of spectograms rather than originally planned as the audio files. I also changed the architecture of the network to now include five convolutional layers that are each followed by a normalization and ReLU layer. The new architecture for my CNN in Matlab code is:<\/p>\n<ol>\n<li>convolution2dLayer<\/li>\n<li>batchNormalizationLayer<\/li>\n<li>reluLayer<\/li>\n<li>maxPooling2dLayer<\/li>\n<li>convolution2dLayer<\/li>\n<li>batchNormalizationLayer<\/li>\n<li>reluLayer<\/li>\n<li>maxPooling2dLayer<\/li>\n<li>convolution2dLayer<\/li>\n<li>batchNormalizationLayer<\/li>\n<li>reluLayer<\/li>\n<li>maxPooling2dLayer<\/li>\n<li>convolution2dLayer<\/li>\n<li>batchNormalizationLayer<\/li>\n<li>reluLayer<\/li>\n<li>convolution2dLayer<\/li>\n<li>batchNormalizationLayer<\/li>\n<li>reluLayer<\/li>\n<li>maxPooling2dLayer<\/li>\n<li>dropoutLayer<\/li>\n<li>fullyConnectedLayer<\/li>\n<li>softmaxLayer<\/li>\n<li>weightedClassificationLayer<\/li>\n<\/ol>\n<p>Here is also a visualization of example spectograms of both normal and abnormal heart sounds from the training data:<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone size-medium wp-image-63\" src=\"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/spectograms1-300x169.jpg\" alt=\"\" width=\"300\" height=\"169\" srcset=\"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/spectograms1-300x169.jpg 300w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/spectograms1-768x432.jpg 768w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/spectograms1-1024x576.jpg 1024w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/spectograms1.jpg 1152w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><\/p>\n<p>I then ran the CNN and it achieved an accuracy of 56.52% in 55 seconds to run. I ran it with 25 epochs.<\/p>\n<p>Here is the graph of the training session:<\/p>\n<p><img decoding=\"async\" loading=\"lazy\" class=\"alignnone size-medium wp-image-64\" src=\"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/cnn_training1-300x153.png\" alt=\"\" width=\"300\" height=\"153\" srcset=\"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/cnn_training1-300x153.png 300w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/cnn_training1-768x393.png 768w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/cnn_training1-1024x523.png 1024w, http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-content\/uploads\/sites\/19\/2019\/03\/cnn_training1.png 1710w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/><\/p>\n<p>Improvements have to be made to the preprocessing of the data as well as finding a more optimal architecture for this data in the next week.<\/p>\n<p>I am currently on schedule for the week.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>For this week, I began classifying the heart sound training data using a CNN. I researched that CNN&#8217;s aren&#8217;t very optimal to use for audio files so instead I am taking the spectograms of the data and then training the network on the images of spectograms rather than originally planned as the audio files. I &hellip; <\/p>\n<p class=\"link-more\"><a href=\"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/2019\/03\/09\/ryan-week-4-status-report\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Ryan &#8211; Week 4 Status Report&#8221;<\/span><\/a><\/p>\n","protected":false},"author":53,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"_links":{"self":[{"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/posts\/61"}],"collection":[{"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/users\/53"}],"replies":[{"embeddable":true,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/comments?post=61"}],"version-history":[{"count":1,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/posts\/61\/revisions"}],"predecessor-version":[{"id":65,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/posts\/61\/revisions\/65"}],"wp:attachment":[{"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/media?parent=61"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/categories?post=61"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/course.ece.cmu.edu\/~ece500\/projects\/s19-teama4\/wp-json\/wp\/v2\/tags?post=61"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}