data <- gdata::read.xls("EDU408saID.xls")

data <- data[,-1]
data <- data[,-5]
data <- data[,-6]
attach(data)
set.seed(77)

ind <- sample(2, nrow(data), replace = TRUE, prob = c(0.7, 0.3))
train <- data[ind==1,]
test <- data[ind==2,]
test_y <- data[ind==2,"GradePASS"]

logistic_model <- glm(GradePASS ~ videos,data = train,family = "binomial")

summary(logistic_model)

logistic_probs <- predict(logistic_model, test, type = "response")

head(logistic_probs)

fix(logistic_probs)

sort_probs <- sort(logistic_probs)

logistic_pred_y = rep(0, length(test_y)) #stavimo da su sve vrijednosti FAIL

logistic_pred_y[logistic_probs > 0.5] = 1 #gdje je >0.5 stavljamo PASS

table(logistic_pred_y, test_y) # matrica konfuzije

mean(logistic_pred_y != test_y)