'How to plot graph of training error rate in KNN
I have the following code:
all_accuracies <- c(accs1, accs2)
all_accuracies
kvalues <- c(1,3,5,7,11)
all_data <- data.frame(kvalues,all_accuracies)
all_data
library(ggplot2)
crossData <- data.frame((k = 1/all_data$kvalues), Accuracy_rate = all_data$all_accuracies)
crossData[1:7] <- "Training Data"
crossData[8:14] <- "Cross Validated Data"
ggplot(crossData, aes(x=k, y=Accuracy_rate)) + geom_line() + geom_point() + labs(x="1/k", y="Accuracy Rate")
On here values of crossData from 1:7 are of normal training data and values from 8:14 are of cross-validated training data.
Basically here, accs1 and accs2 contain accuracy rate of training set and accuracy rate of cross validated training set.
I want to show the accuracy rate over training set and cross-validated accuracy rate of the set for each value of k. I want to plot a graph that looks similar to:
How can I code to draw this plot? I'm having troubles with it.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|

