Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
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
| calculatePforWelchtTest = function(s) { | |
| t.value = (s['mean', 'x'] - s['mean', 'y'] - 0) / | |
| (sqrt(s['var', 'x']/s['#samples', 'x'] + | |
| s['var', 'y']/s['#samples', 'y'] ) | |
| ) | |
| r = (s['var', 'x']/s['#samples', 'x'] + | |
| s['var', 'y']/s['#samples', 'y'] )^2 / | |
| ( (s['var', 'x']/s['#samples', 'x'])^2/(s['#samples', 'x']-1) + | |
| (s['var', 'y']/s['#samples', 'y'])^2/(s['#samples', 'y']-1) |
halilbilgin
/ variableImportance.R
Created
November 12, 2017 16:51
Variable Importance function for deep learning
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
| #This function assumes that there is at least one hidden layer. | |
| variableImportance = function(modelTf) { | |
| numberOfWeights = c(dim(modelTf$weight$W1)[2], dim(modelTf$weight$W2)[2], dim(modelTf$weight$W2)[1]) | |
| Qik = matrix(0, numberOfWeights[1], numberOfWeights[3]) | |
| sumWj = array(0, numberOfWeights[2]); | |
| sumWk = array(0, numberOfWeights[3]); | |
| for(j in 1:numberOfWeights[2]){ | |
| sumWj[j] = sum(abs(modelTf$weight$W1[,j])) |
halilbilgin
/ MontyHall.java
Created
October 21, 2017 18:57
Solve Monty Hall Problem using Monte Carlo simulations
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
| public class MontyHall { | |
| int usersChoice; | |
| int doorWithTheCar; | |
| } |
halilbilgin
/ fitNuSVR.R
Last active
October 27, 2017 06:26
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
| nuSVRGrid <- expand.grid( | |
| gamma = c(0.0000305176, 0.000122070, 0.000488281, 0.00195313, 0.0078125, 0.03125, 0.125, 0.5, 2, 8, 32), | |
| nu = c(0.01,0.05,0.1,0.15,0.2,0.25,0.3,0.35,0.4,0.45,0.5, | |
| 0.55,0.6,0.65,0.7,0.75,0.8,0.85,0.9,0.95,1.0), | |
| C = c(0.0000305176,0.000122070,0.000488281,0.00195313,0.0078125, | |
| 0.03125,0.125,0.5,2,8,32) | |
| ) | |
| library(doParallel) | |
| registerDoParallel(cores=10) |
halilbilgin
/ caretNuSVRModelInfo.R
Last active
October 27, 2017 06:26
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
| modelInfo <- list(label = "nu-SVR with Radial Basis Function Kernel", | |
| library = "e1071", | |
| type = c("Regression"), | |
| parameters = data.frame(parameter = c("nu", "C", "gamma"), | |
| class = c("numeric", "numeric", "numeric"), | |
| label = c("Nu", "Cost", "Gamma")), | |
| loop = NULL, | |
| grid=NULL, | |
| fit = function(x, y, wts, param, lev, last, classProbs, ...) { | |
| if(any(names(list(...)) == "prob.model") | is.numeric(y)) { |
halilbilgin
/ trainingTest
Last active
June 14, 2017 01:25
split data into training and test sets without data leakage
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
| trTest <- function(dataset, p = 0.75, scaleData='min-max', colsToBeScaled = c()) { | |
| dataset[, 'train'] <- ifelse(runif(nrow(dataset)) <= p, 1, 0) | |
| trData <- dataset[dataset[, 'train'] == 1,] | |
| tData <- dataset[dataset[, 'train'] == 0,] | |
| trData[,'train'] <- NULL | |
| tData[,'train'] <- NULL | |
| if(scaleData != FALSE) { |
halilbilgin
/ crossValidation
Last active
June 14, 2017 01:23
do cross validation without data leakage
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
| crossValidation <- function(dataset, k = 10, scaleData='min-max', colsToBeScaled = c()) { | |
| require(caret) | |
| #create fold ids | |
| folds <- createFolds(as.integer(row.names(dataset)), k = k) | |
| folds <- lapply(folds, function(x) { | |
| scaleData <- get('scaleData') | |
| colsToBeScaled <- get('colsToBeScaled') | |
| trData <- get('dataset')[-c(x), ] | |
| tData <- get('dataset')[c(x), ] |