Some AI algo works better with values between 0 and 1 but it is rare to have data already between 0 and 1.
The following formula will show you how to convert an array of data to a normalized
Normalize = ($inputvalue – $min) / ($max – $min)
0.111 = (2-1)/(10-1)
Denormalized = ($normalizedValue * ($max-$min) + $min)
3 = (0.22222 * (10-1) + 1)
| Input Value | Normalise | Denormalise |
| ($inputvalue – $min) / ($max – $min) | ($normalizedValue * ($max-$min) + $min) | |
| 1 | 0 | 1 |
| 2 | 0.111111111 | 2 |
| 3 | 0.222222222 | 3 |
| 4 | 0.333333333 | 4 |
| 5 | 0.444444444 | 5 |
| 6 | 0.555555556 | 6 |
| 7 | 0.666666667 | 7 |
| 8 | 0.777777778 | 8 |
| 9 | 0.888888889 | 9 |
| 10 | 1 | 10 |
The post How to normalize and denormalize data between 0 and 1 appeared first on Martin Fournier.
