monogenea

# Split train/test with rsample
set.seed(100)
initSplit <- initial_split(X, prop = .9,
                           strata = "Y")
trainSet <- training(initSplit)
testSet <- testing(initSplit)

# Create 5-fold cross-validation, convert to caret class
set.seed(100)
myFolds <- vfold_cv(trainSet, v = 5, repeats = 5,

monogenea

# binary vars
binVars <- which(sapply(X, function(x){all(x %in% 0:1)}))
missingVars <- which(apply(X, 2, function(k){any(is.na(k))}))

monogenea

# binary vars
binVars <- which(sapply(X, function(x){all(x %in% 0:1)}))
missingVars <- which(apply(X, 2, function(k){any(is.na(k))}))

monogenea

# Design recipe
myRec <- recipe(Y ~ ., data = trainSet) %>% 
      step_YeoJohnson(all_predictors(), -binVars) %>% 
      step_center(all_predictors(), -binVars) %>% 
      step_scale(all_predictors(), -binVars) %>% 
      step_meanimpute(missingVars)

monogenea

# simple PCA, plot
pcaRec <- myRec %>% 
      step_pca(all_predictors())

myPCA <- prep(pcaRec, training = trainSet, retain = T) %>% 
      juice()
colGrad <- trainSet$Y/100 # add color

plot(myPCA$PC1, myPCA$PC2,
     col = rgb(1 - colGrad, 0, colGrad,.5),

monogenea

# Train
doMC::registerDoMC(10)
knnMod <- train(myRec, data = trainSet,
                method = "knn",
                tuneGrid = data.frame(k = seq(5, 25, by = 4)),
                trControl = ctrl)

enetMod <- train(myRec, data = trainSet,
                 method = "glmnet",
                 tuneGrid = expand.grid(alpha = seq(0, 1, length.out = 5),

monogenea

bwplot(resamples(modelList),
       metric = "RMSE")

monogenea

# Validate on test set with ensemble
allPreds <- sapply(modelList, predict, newdata = testSet)
ensemblePred <- rowSums(allPreds) / length(modelList)

# Plot predicted vs. observed; create PNG
plot(ensemblePred, testSet$Y,
     xlim = c(0,100), ylim = c(0,100),
     xlab = "Predicted", ylab = "Observed",
     pch = 16, col = rgb(0, 0, 0, .25))
abline(a=0, b=1)

monogenea

##### Process image #####
library(keras)
library(EBImage)
library(stringr)
library(pbapply)

secondCat <- readImage("train/cat.1.jpg")
display(secondCat)

monogenea

# Set image size
width <- 50
height <- 50

extract_feature <- function(dir_path, width, height, labelsExist = T) {
      img_size <- width * height
      
      ## List images in path
      images_names <- list.files(dir_path)