Created
April 10, 2013 15:27
-
-
Save ovatsus/5355630 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#r "bin/v40/FSharp.Data.dll" | |
open FSharp.Data.Csv | |
open FSharp.Data.Csv.Extensions | |
let map f (csvFile:CsvFile) = | |
let headers = csvFile.Headers |> f |> String.concat "," | |
let data = csvFile.Data |> Seq.map (fun row -> row.Columns |> f |> String.concat ",") |> String.concat "\n" | |
CsvFile.Parse(headers + "\n" + data) | |
let filter f (csvFile:CsvFile) = | |
let headers = csvFile.Headers |> String.concat "," | |
let data = csvFile.Data |> Seq.map (fun row -> row.Columns) |> Seq.filter f |> Seq.map (String.concat ",") |> String.concat "\n" | |
CsvFile.Parse(headers + "\n" + data) | |
type CsvAtomicOperationOnColumn = { Col : int; Op : float -> bool } | |
type CsvOperationOnColumn = | |
| Atomic of CsvAtomicOperationOnColumn | |
| Or of CsvOperationOnColumn * CsvOperationOnColumn | |
| And of CsvOperationOnColumn * CsvOperationOnColumn | |
static member (.&&) (op1, op2) = And(op1, op2) | |
static member (.||) (op1, op2) = Or(op1, op2) | |
member x.Evaluate(row:string[]) = | |
match x with | |
| Atomic { Col = index; Op = op } -> row.[index].AsFloat() |> op | |
| Or(op1, op2) -> op1.Evaluate(row) || op2.Evaluate(row) | |
| And(op1, op2) -> op1.Evaluate(row) && op2.Evaluate(row) | |
type CsvColumn = | |
{ File : CsvFile; Column : int } | |
static member (.>) (c:CsvColumn, value) = Atomic { Col = c.Column; Op = ((<) value) } | |
static member (.<) (c:CsvColumn, value) = Atomic { Col = c.Column; Op = ((>) value) } | |
static member (.=) (c:CsvColumn, value) = Atomic { Col = c.Column; Op = ((=) value) } | |
member x.Filter(row:CsvRow) = row.Columns.[x.Column] | |
member x.Data = | |
let t:CsvFile = x.File.[x] | |
t.Data |> Seq.map (fun row -> (row.Columns |> Seq.exactlyOne).AsFloat()) | |
and CsvFile with | |
member x.Ozone = { File = x; Column = 0 } | |
member x.``Solar.R`` = { File = x; Column = 1 } | |
member x.Wind = { File = x; Column = 2 } | |
member x.Temp = { File = x; Column = 3 } | |
member x.Month = { File = x; Column = 4 } | |
member x.Day = { File = x; Column = 5 } | |
member x.Item with get({ File = x; Column = index }) = x |> map (fun row -> row.[index] |> Seq.singleton) | |
member x.Item with get([<System.ParamArrayAttribute>]indices : CsvColumn[]) = x |> map (fun row -> row |> Seq.mapi (fun i y -> y, indices |> Seq.tryFind ((=) { File = x; Column = i })) |> Seq.filter (fun (_, i) -> i.IsSome) |> Seq.map fst) | |
member x.Item with get(op:CsvOperationOnColumn) = x |> filter (fun row -> op.Evaluate row) | |
let csv = CsvFile.Load("http://faculty.washington.edu/heagerty/Books/Biostatistics/DATA/ozone.csv") | |
//How many missing values are in the Ozone column of this data frame? | |
open System | |
csv.Ozone.Data |> Seq.countBy Double.IsNaN | |
//What is the mean of the Ozone column in this dataset? Exclude missing values (coded as NA) from this calculation. | |
csv.Ozone.Data |> Seq.filter (not << Double.IsNaN) |> Seq.average | |
//Extract the subset of rows of the data frame where Ozone values are above 31 and Temp values are above 90. What is the mean of Solar.R in this subset? | |
csv.[(csv.Ozone .> 31.0) .&& (csv.Temp .> 90.0)].``Solar.R``.Data |> Seq.average | |
//What is the mean of "Temp" when "Month" is equal to 6? | |
csv.[csv.Month .= 6.0].Temp.Data |> Seq.average |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment