Back to Questions
853
views7
comments
Iterating floats using List
f#

Hi,

I'm new to F# but have used C# for several years.

I'm using List.filter to iterate through a series of floats, like so:

1

let lst = List.filter (fun x -> x = x) [100000.00 .. 999999.00]

The reason for the post is the warning I receive regarding the decimal places in the range.

Floating point ranges are experimental and may be deprecated in a future release. Consider using an integer range instead, converting to floating point using 'float' or 'float32'.

I need to treat x as a float, how can I achieve this using a integer range.

Any help would be much appreciated.

Rich

.

Hi, there. Are you really filtering the list? I thought you were going to iterate the list and apply something/some actions to each element. If so, then List.iter is your option. Since floating point range is not a recommendation for the time being, you could first apply List.map to an integer list then List.iter like:

1

[100000 .. 999999] |> List.map (fun x -> float x) |> List.iter (fun x -> printfn "%f" x)

I think what you want is to transform the list, right?

Hope that helps!

By on 7/29/2008 7:49 PM ()

Hi Allen,

I am actually wanting to filter the list. I'm intrigued by the use of List.map to List.iter, doesn't it double the overhead by going through the list twice? Or, is F# clever enough to apply both functions on one pass?

Rich

By on 7/30/2008 12:49 AM ()

Hi Rich,

A couple of things:
- you shouldn't assume that F# will be smart enough to go through the list once, some things might get optimized but I believe this code will take two passes,
- doing an equality check on floats is not considered very safe
- it's also a bit weird to write List.filter (fun x -> float x = float x) as that doesn't accomplish anything; it just checks whether everyone in the list is equal to itself ...

Cheers, J

By on 7/30/2008 1:06 AM ()

Hi Jurgen,

1. I thought that would the case, which is why I didn't write it that way in the first place.

2. Is there some way I can check equality of two decimalised values that is safe?

3. The anonymous function provided was just an example. The actual function is below:

1

(fun x -> float x = sqSum (float x))

Thanks again for your help.

Rich

By on 7/30/2008 4:15 AM ()

Hi Rich & Jurgen

Fun fact for the day:

1

List.filter (fun x -> float x = float x)

does do something: if the input list contains, say, float or float32 values then it filters the list for NaN values, (NaN = NaN returns false under IEEE semantics)...

Disturbing but true :-)

don

By on 7/30/2008 4:51 PM ()

How about the following

1

let lst = List.filter (fun x -> float x = float x) [100000 .. 999999]
By on 7/29/2008 2:34 AM ()

I'm still finding my feet, I knew it'd be really simple. Thank you for your help.

Rich

By on 7/29/2008 2:48 AM ()

    Topic tags

    IntelliFactory Offices Copyright (c) 2011-2012 IntelliFactory. All rights reserved.
    Home | Products | Consulting | Trainings | Blogs | Jobs | Contact Us | Terms of Use | Privacy Policy | Cookie Policy     		Built with WebSharper