Hi All,

  Can anyone  guide me what is the problem  with below code as it throws  error.

import std.stdio, mir.math.common: approxEqual;
static immutable x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9];
static immutable y = [1, 3, 2, 5, 7, 8, 8, 9, 10, 12];
auto params = x.simpleLinearRegression(y);
writeln(params);

On Monday, 30 March 2020 at 17:21:57 UTC, Vino wrote:
> Hi All,
>
>   Can anyone  guide me what is the problem  with below code as it throws  error.
>
> import std.stdio, mir.math.common: approxEqual;
> static immutable x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9];
> static immutable y = [1, 3, 2, 5, 7, 8, 8, 9, 10, 12];
> auto params = x.simpleLinearRegression(y);
> writeln(params);

This should be in the Learn forum.

As far as I can tell, the problem is that x and y are both type immutable(int[]), which it cannot handle. I tried just turning them into slices, but that wasn't sufficient, I had to ensure they were double as well (float would also work). Note that the example in the document uses a y variable that is a floating point type.

/+dub.sdl:
dependency "mir-algorithm" version="~>3.7.25"
+/
import std.stdio : writeln;
import mir.math.stat : simpleLinearRegression;
import mir.ndslice : sliced;

void main() {
    auto x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9].sliced!double;
    auto y = [1, 3, 2, 5, 7, 8, 8, 9, 10, 12].sliced!double;
    auto params = x.simpleLinearRegression(y);
    writeln(params);
}

On Monday, 30 March 2020 at 17:47:19 UTC, jmh530 wrote:
> 
> [snip]

The relevant function signature is

@safe sumType!YRange[2] simpleLinearRegression(XRange, YRange)(XRange x, YRange y)
if (isInputRange!XRange && isInputRange!YRange && !(isArray!XRange && isArray!YRange) && isFloatingPoint!(sumType!YRange));

and when you pass an int[] for y, then it determines that int is not a floating point and you get an error about not matching the signature, nothing about why you're not matching it.

A good example of where Atila's concepts library would be helpful...

On Monday, 30 March 2020 at 17:21:57 UTC, Vino wrote:
> Hi All,
>
>   Can anyone  guide me what is the problem  with below code as it throws  error.
>
> import std.stdio, mir.math.common: approxEqual;
> static immutable x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9];
> static immutable y = [1, 3, 2, 5, 7, 8, 8, 9, 10, 12];
> auto params = x.simpleLinearRegression(y);
> writeln(params);

static immutable y = [1.0, 3, 2, 5, 7, 8, 8, 9, 10, 12];

`.0` will change the sum type of y to double.