I ran the following to generate some data (from your tests dir):
> gen_data <- function(N){
+ X <- data.frame(x0 = 1:N, x1 = rnorm(N))
+ y <- 1:N
+ return(list(X, y))
+ }
>
> # Generate test input data
> data <- gen_data(100)
> X <- data[[1]]
> y_vec <- data[[2]]
> y <- data.frame(y = y_vec)
then I tried your example code:
> split_data <- train_test_split(X, y, test_size = 0.25, random_state = 0, shuffle = TRUE)
> scores <- cross_validation(split_data['X_train'], split_data['y_train'], k = 3, shuffle = TRUE, random_state = 0)
I received this error after the last line of code pasted above:
Error in cross_validation(split_data["X_train"], split_data["y_train"], :
TypeError: X must be a dataframe or an atomic vector