Merge pull request #652 from joshua-whittemore/add-module-to-download-iris-dataset

Add module to make iris dataset available.

NEWS.md

@@ -9,6 +9,7 @@
 * New "performance tips" [section of the docs](https://github.com/FluxML/Flux.jl/pull/615).
 * The training loop is [now more readable](https://github.com/FluxML/Flux.jl/pull/651) and better shows how to use the lower-level APIs.
 * New [AlphaDropout](https://github.com/FluxML/Flux.jl/pull/656).
+* [Data.Iris](https://github.com/FluxML/Flux.jl/pull/652) makes Fisher's Iris dataset available with `Iris.labels` and `Iris.features`.
 * New [InstanceNorm](https://github.com/FluxML/Flux.jl/pull/634), as popularized by [Instance Normalization: The Missing Ingredient for Fast Stylization](https://arxiv.org/abs/1607.08022).
 
 AD Changes:

Project.toml

@@ -6,6 +6,7 @@ AbstractTrees = "1520ce14-60c1-5f80-bbc7-55ef81b5835c"
 Adapt = "79e6a3ab-5dfb-504d-930d-738a2a938a0e"
 CodecZlib = "944b1d66-785c-5afd-91f1-9de20f533193"
 Colors = "5ae59095-9a9b-59fe-a467-6f913c188581"
+DelimitedFiles = "8bb1440f-4735-579b-a4ab-409b98df4dab"
 Juno = "e5e0dc1b-0480-54bc-9374-aad01c23163d"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
 MacroTools = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"

src/data/Data.jl

@@ -39,4 +39,7 @@ include("tree.jl")
 include("sentiment.jl")
 using .Sentiment
 
+include("iris.jl")
+export Iris
+
 end

src/data/iris.jl

@@ -0,0 +1,88 @@
+
+"""
+
+    Iris
+
+Fisher's classic iris dataset.
+
+Measurements from 3 different species of iris: setosa, versicolor and 
+virginica.  There are 50 examples of each species.
+
+There are 4 measurements for each example: sepal length, sepal width, petal 
+length and petal width.  The measurements are in centimeters.
+
+The module retrieves the data from the [UCI Machine Learning Repository](https://archive.ics.uci.edu/ml/datasets/iris).
+
+"""
+module Iris
+
+using DelimitedFiles
+using ..Data: deps, download_and_verify
+
+const cache_prefix = ""
+
+# Uncomment if the iris.data file is cached to cache.julialang.org.
+# const cache_prefix = "https://cache.julialang.org/"
+
+function load()
+    isfile(deps("iris.data")) && return
+
+    @info "Downloading iris dataset."
+    download_and_verify("$(cache_prefix)https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data",
+                        deps("iris.data"),
+                        "6f608b71a7317216319b4d27b4d9bc84e6abd734eda7872b71a458569e2656c0")
+end
+
+"""
+
+    labels()
+
+Get the labels of the iris dataset, a 150 element array of strings listing the 
+species of each example.
+
+```jldoctest
+julia> labels = Flux.Data.Iris.labels();
+
+julia> summary(labels)
+"150-element Array{String,1}"
+
+julia> labels[1]
+"Iris-setosa"
+```
+"""
+function labels()
+    load()
+    iris = readdlm(deps("iris.data"), ',')
+    Vector{String}(iris[1:end, end])
+end
+
+"""
+
+    features()
+
+Get the features of the iris dataset.  This is a 4x150 matrix of Float64 
+elements.  It has a row for each feature (sepal length, sepal width, 
+petal length, petal width) and a column for each example.
+
+```jldoctest
+julia> features = Flux.Data.Iris.features();
+
+julia> summary(features)
+"4×150 Array{Float64,2}"
+
+julia> features[:, 1]
+4-element Array{Float64,1}:
+ 5.1
+ 3.5
+ 1.4
+ 0.2
+```
+"""
+function features()
+    load()
+    iris = readdlm(deps("iris.data"), ',')
+    Matrix{Float64}(iris[1:end, 1:4]')
+end
+end
+
+

test/data.jl

@@ -14,3 +14,9 @@ using Test
 @test FashionMNIST.labels() isa Vector{Int64}
 
 @test Data.Sentiment.train() isa Vector{Data.Tree{Any}}
+
+@test Iris.features() isa Matrix
+@test size(Iris.features()) == (4,150)
+
+@test Iris.labels() isa Vector{String}
+@test size(Iris.labels()) == (150,)