ConvertDoubleToInt modified by KKozlov

KKozlov — Wed, 03 Feb 2021 06:35:52 -0000

--- v1
+++ v2
@@ -4,7 +4,7 @@

 DEEP method operates on floating point parameters. On the other hand many real problems are formulated so that the parameters are integer numbers, e.g. indices of some sort. 

-Particularly, this is the case for the problem of constructing the molecule with desired properties. 
+### Conversion by sorting

 The following procedure is used to produce the integer vector of parameters from the floating point vector constructed by DEEP method. 

@@ -34,3 +34,9 @@
     i[2] = 1
     i[3] = 3

+### Conversion by special rounding
+
+Taken from Uher et al., 2016
+```
+x_int = (1+999*x_double)/500
+```

ConvertDoubleToInt modified by KKozlov

KKozlov — Wed, 11 Feb 2015 12:39:35 -0000

Last modified 2 years ago

Finding Integer Valued Parameters with DEEP

DEEP method operates on floating point parameters. On the other hand many real problems are formulated so that the parameters are integer numbers, e.g. indices of some sort.

Particularly, this is the case for the problem of constructing the molecule with desired properties.

The following procedure is used to produce the integer vector of parameters from the floating point vector constructed by DEEP method.

The values are sorted in ascending order.
The index of the parameter in the floating point array becomes the value of the parameter in the integer array.

Floating point array;

a[0] = 0.3
a[1] = 0.5
a[2] = 0.1
a[3] = 0.8

Sorted array:

b[0] = a[2] = 0.1
b[1] = a[0] = 0.3
b[2] = a[1] = 0.5
b[3] = a[3] = 0.8

Integer array:

i[0] = 2
i[1] = 0
i[2] = 1
i[3] = 3

Recent changes to ConvertDoubleToInt

ConvertDoubleToInt modified by KKozlov

ConvertDoubleToInt modified by KKozlov

Finding Integer Valued Parameters with DEEP