Create run_analysis.R

run_analysis.R

@@ -0,0 +1,33 @@
+subjectTrainData = read.table("./train/subject_train.txt")
+subjectTestData = read.table("./test/subject_test.txt")
+
+xTrainData = read.table("./train/X_train.txt")
+yTrainData = read.table("./train/y_train.txt")
+
+xTestData = read.table("./test/X_test.txt")
+yTestData = read.table("./test/y_test.txt")
+
+featuresData = read.table("features.txt", stringsAsFactors=FALSE)
+activityLabels = read.table("activity_labels.txt", stringsAsFactors=FALSE)
+
+sum(is.na(xTrainData))
+sum(is.na(yTrainData))
+sum(is.na(xTestData))
+sum(is.na(yTestData))
+
+xData = rbind(xTrainData, xTestData)
+yData = rbind(yTrainData, yTestData)
+subjectData = rbind(subjectTrainData, subjectTestData)
+
+meanStdCols = grep("(-mean\\(\\))|(-std\\(\\))", featuresData$V2)
+selected.x.data = xData[, meanStdCols]
+
+colnames(selected.x.data) = featuresData$V2[meanStdCols]
+colnames(subjectData) = c("SubjectId")
+colnames(yData) = c("Activity")
+
+data.full = cbind(subjectData, selected.x.data, yData)
+data.final = merge(data.full, activityLabels, by.x = "Activity", by.y = "V1")[, -1]
+colnames(data.final)[68] = "Activity"
+
+agg.data.final = aggregate(data.final[, 2:67], by = data.final[,c("SubjectId", "Activity")], mean)