Normalized to standard in Rstudio
1
0
Entering edit mode
11 months ago

For Neural Network analysis, I normalized my data using the following function in Rstudio:

normalize <- function(x) { return ((x - min(x)) / (max(x) - min(x))) }
weekends1 <- as.data.frame(lapply(weekends1, normalize))


Now that I have my desired result, how can I present my data as a standard form not a normalized one?

r • 324 views
0
Entering edit mode

It's not clear what you're asking. You've normalized but want to show the original data? More detail is necessary, or better yet, a clear, reproducible example to show exactly what you want and the issue you're having.

0
Entering edit mode

the output is a plot with normalized values, however, I need to plot the result of neural network analysis with the initial values. For the neural network analysis, values should be normalized, but for my results, I need to show it using initial values. ( I am using a neural network to predict indoor temperature, Currently, I have the prediction in the range of 0-1, but I need to convert them to Celsius as it was initially to plot my results.

0
Entering edit mode

I may be missing something, but can you not just plot the original data along-side it then?

0
Entering edit mode

I just need to present my plot with degree celsius not a normalized value that is meaningless for audiences

1
Entering edit mode
11 months ago

Normalization:

max = 30C
min = 10C

norm(x) = (x - min(x)) / (max(x) - min(x)
norm(22C) = (22 - 10) / (30 - 10)
norm(22C) = 0.6


Normalized conversion to C:

0.6 = (x - 10) / (30 - 10)
12 = x - 10
22 = x