How to simplify the code using Apply Function - r

Question

i have this script:
library(plyr)
library(gstat)
library(sp)
library(dplyr)
library(ggplot2)
library(scales)
a<-c(10,20,30,40,50,60,70,80,90,100)
b<-c(15,25,35,45,55,65,75,85,95,105)
x<-rep(a,3)
y<-rep(b,3)
E<-sample(30)
freq<-rep(c(100,200,300),10)
data<-data.frame(x,y,freq,E)
data<-arrange(data,x,y,freq)
df <- ddply(data,"freq", function (h){
dim_h<-length(h$x)
perc_max <- 0.9
perc_min <- 0.8
u<-round((seq(perc_max,perc_min,by=-0.1))*dim_h)
dim_u<-length(u)
perc_punti<- percent(seq(perc_max,perc_min,by=-0.1))
for (i in 1:dim_u)
{ t<-u[i]
time[i]<-system.time(
for (j in 1:2)
{
df_tass <- sample_n(h, t)
df_residuo <- slice(h,-as.numeric(rownames(df_tass)))
coordinates(df_tass)= ~x + y
x.range <- range(h$x)
y.range <- range(h$y)
grid <- expand.grid(x = seq(from = x.range[1], to = x.range[2], by = 1), y = seq(from = y.range[1],
to = y.range[2], by = 1))
coordinates(grid) <- ~x + y
gridded(grid) <- TRUE
nearest = krige(E ~ 1, df_tass, grid, nmax = 1)
nearest_df<-as.data.frame(nearest)
names(nearest_df) <- c("x", "y", "E")
#Error of prediction
df_pred <- inner_join(nearest_df[1:3],select(df_residuo,x,y,E),by=c("x","y"))
names(df_pred) <- c("x", "y", "E_pred","E")
sqm[j] <- mean((df_pred[,4]-df_pred[,3])^2)
})[3]
sqmm[i]<-mean(sqm)
}
df_finale<-data.frame(sqmm,time,perc_punti)
})
df
I measured in several points of coordinates (x,y) the value of the electromagnetic field (E value) at different frequencies (freq value). For each frequency value, I use once 90% of points and once 80% (with the for loop with l) to interpolate the value of the electromagnetic field (E) inside grid with Nearest Neighbour Interpolation (krige function); and i repeat this 2 times. The remaining points will then be used to calculate the prediction error. I hope it's clear.
This script above is a simplified case. Unfortunately, in my case the script takes too long for the two for-loops implemented.
I want to ask if it's possible to simplify the code in some way, for instance by using the apply function family. Thanks.
Reply #clemlaflemme ok it works! thanks... now i have a little proble with the final dataframe, it looks like this:
freq 1 2
1 100 121.00 338.00
2 100 0.47 0.85
3 200 81.00 462.50
4 200 0.74 0.73
5 300 36.00 234.00
6 300 0.82 0.76
but i want something like this:
freq sqmm time
1 100 121.0 0.47
2 100 338.0 0.85
3 200 81.0 0.74
4 200 462.5 0.73
5 300 36.0 0.82
6 300 234.0 0.76
how can i do that??