Skip to content

Commit 24ccd37

Browse files
rustdrustd
committed
Clustering_Iris sample updated
1 parent 058b915 commit 24ccd37

File tree

7 files changed

+621
-0
lines changed

7 files changed

+621
-0
lines changed

‎.gitattributes

+63
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,63 @@
1+
###############################################################################
2+
# Set default behavior to automatically normalize line endings.
3+
###############################################################################
4+
* text=auto
5+
6+
###############################################################################
7+
# Set default behavior for command prompt diff.
8+
#
9+
# This is need for earlier builds of msysgit that does not have it on by
10+
# default for csharp files.
11+
# Note: This is only used by command line
12+
###############################################################################
13+
#*.cs diff=csharp
14+
15+
###############################################################################
16+
# Set the merge driver for project and solution files
17+
#
18+
# Merging from the command prompt will add diff markers to the files if there
19+
# are conflicts (Merging from VS is not affected by the settings below, in VS
20+
# the diff markers are never inserted). Diff markers may cause the following
21+
# file extensions to fail to load in VS. An alternative would be to treat
22+
# these files as binary and thus will always conflict and require user
23+
# intervention with every merge. To do so, just uncomment the entries below
24+
###############################################################################
25+
#*.sln merge=binary
26+
#*.csproj merge=binary
27+
#*.vbproj merge=binary
28+
#*.vcxproj merge=binary
29+
#*.vcproj merge=binary
30+
#*.dbproj merge=binary
31+
#*.fsproj merge=binary
32+
#*.lsproj merge=binary
33+
#*.wixproj merge=binary
34+
#*.modelproj merge=binary
35+
#*.sqlproj merge=binary
36+
#*.wwaproj merge=binary
37+
38+
###############################################################################
39+
# behavior for image files
40+
#
41+
# image files are treated as binary by default.
42+
###############################################################################
43+
#*.jpg binary
44+
#*.png binary
45+
#*.gif binary
46+
47+
###############################################################################
48+
# diff behavior for common document formats
49+
#
50+
# Convert binary document formats to text before diffing them. This feature
51+
# is only available from the command line. Turn it on by uncommenting the
52+
# entries below.
53+
###############################################################################
54+
#*.doc diff=astextplain
55+
#*.DOC diff=astextplain
56+
#*.docx diff=astextplain
57+
#*.DOCX diff=astextplain
58+
#*.dot diff=astextplain
59+
#*.DOT diff=astextplain
60+
#*.pdf diff=astextplain
61+
#*.PDF diff=astextplain
62+
#*.rtf diff=astextplain
63+
#*.RTF diff=astextplain

‎samples/csharp/getting-started/Clustering_Iris/Clustering_Iris.sln

+25
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
2+
Microsoft Visual Studio Solution File, Format Version 12.00
3+
# Visual Studio 15
4+
VisualStudioVersion = 15.0.28010.2046
5+
MinimumVisualStudioVersion = 10.0.40219.1
6+
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Clustering_Iris", "Clustering_Iris\Clustering_Iris.csproj", "{E730C84B-0F03-4C0C-9B22-68130091C900}"
7+
EndProject
8+
Global
9+
GlobalSection(SolutionConfigurationPlatforms) = preSolution
10+
Debug|Any CPU = Debug|Any CPU
11+
Release|Any CPU = Release|Any CPU
12+
EndGlobalSection
13+
GlobalSection(ProjectConfigurationPlatforms) = postSolution
14+
{E730C84B-0F03-4C0C-9B22-68130091C900}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
15+
{E730C84B-0F03-4C0C-9B22-68130091C900}.Debug|Any CPU.Build.0 = Debug|Any CPU
16+
{E730C84B-0F03-4C0C-9B22-68130091C900}.Release|Any CPU.ActiveCfg = Release|Any CPU
17+
{E730C84B-0F03-4C0C-9B22-68130091C900}.Release|Any CPU.Build.0 = Release|Any CPU
18+
EndGlobalSection
19+
GlobalSection(SolutionProperties) = preSolution
20+
HideSolutionNode = FALSE
21+
EndGlobalSection
22+
GlobalSection(ExtensibilityGlobals) = postSolution
23+
SolutionGuid = {740BE814-1300-4D85-B982-D0384949FC5D}
24+
EndGlobalSection
25+
EndGlobal

‎samples/csharp/getting-started/Clustering_Iris/Clustering_Iris/Clustering_Iris.csproj

+18
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
<Project Sdk="Microsoft.NET.Sdk">
2+
3+
<PropertyGroup>
4+
<OutputType>Exe</OutputType>
5+
<TargetFramework>netcoreapp2.1</TargetFramework>
6+
</PropertyGroup>
7+
8+
<ItemGroup>
9+
<PackageReference Include="Microsoft.ML" Version="0.6.0" />
10+
</ItemGroup>
11+
12+
<ItemGroup>
13+
<None Update="datasets\iris-full.txt">
14+
<CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
15+
</None>
16+
</ItemGroup>
17+
18+
</Project>

‎samples/csharp/getting-started/Clustering_Iris/Clustering_Iris/Program.cs

+151
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,151 @@
1+
using Microsoft.ML;
2+
using Microsoft.ML.Core.Data;
3+
using Microsoft.ML.Runtime.Api;
4+
using Microsoft.ML.Runtime.Data;
5+
using Microsoft.ML.Runtime.KMeans;
6+
using Microsoft.ML.Runtime.Learners;
7+
using System;
8+
using System.IO;
9+
10+
namespace Clustering_Iris
11+
{
12+
internal static class Program
13+
{
14+
private static string AppPath => Path.GetDirectoryName(Environment.GetCommandLineArgs()[0]);
15+
private static string DataPath => Path.Combine(AppPath, "datasets", "iris-full.txt");
16+
private static string ModelPath => Path.Combine(AppPath, "IrisModel.zip");
17+
18+
19+
private static void Main(string[] args)
20+
{
21+
// Create ML.NET context/environment
22+
using (var env = new LocalEnvironment())
23+
{
24+
// Create DataReader with data schema mapped to file's columns
25+
var reader = new TextLoader(env,
26+
new TextLoader.Arguments()
27+
{
28+
Separator = "\t",
29+
HasHeader = true,
30+
Column = new[]
31+
{
32+
new TextLoader.Column("Label", DataKind.R4, 0),
33+
new TextLoader.Column("SepalLength", DataKind.R4, 1),
34+
new TextLoader.Column("SepalWidth", DataKind.R4, 2), new TextLoader.Column("SepalWidth", DataKind.R4, 2),
35+
new TextLoader.Column("PetalLength", DataKind.R4, 3),
36+
new TextLoader.Column("PetalWidth", DataKind.R4, 4),
37+
38+
}
39+
});
40+
//Load training data
41+
IDataView trainingDataView = reader.Read(new MultiFileSource(DataPath));
42+
43+
// Transform your data and add a learner
44+
// Add a learning algorithm to the pipeline. e.g.(What are characteristics of iris is this?)
45+
// Convert the Label back into original text (after converting to number in step 3)
46+
var pipeline = new ConcatEstimator(env, "Features", "SepalLength", "SepalWidth", "PetalLength", "PetalWidth")
47+
.Append(new KMeansPlusPlusTrainer(env, "Features",clustersCount:3));
48+
49+
// Create and train the model
50+
Console.WriteLine("=============== Create and Train the Model ===============");
51+
52+
var model = pipeline.Fit(trainingDataView);
53+
54+
Console.WriteLine("=============== End of training ===============");
55+
Console.WriteLine();
56+
57+
// Test with one sample text
58+
var sampleIrisData = new IrisData()
59+
{
60+
SepalLength = 3.3f,
61+
SepalWidth = 1.6f,
62+
PetalLength = 0.2f,
63+
PetalWidth = 5.1f,
64+
};
65+
66+
var prediction = model.MakePredictionFunction<IrisData, IrisPrediction>(env).Predict(
67+
sampleIrisData);
68+
69+
Console.WriteLine($"Clusters assigned for setosa flowers:"+prediction.SelectedClusterId);
70+
// Save model to .ZIP file
71+
SaveModelAsFile(env, model);
72+
73+
// Predict again but now testing the model loading from the .ZIP file
74+
PredictWithModelLoadedFromFile(sampleIrisData);
75+
76+
Console.WriteLine("=============== End of process, hit any key to finish ===============");
77+
Console.ReadKey();
78+
}
79+
80+
81+
}
82+
83+
private static void SaveModelAsFile(LocalEnvironment env, TransformerChain<ClusteringPredictionTransformer<KMeansPredictor>> model)
84+
{
85+
using (var fs = new FileStream(ModelPath, FileMode.Create, FileAccess.Write, FileShare.Write))
86+
model.SaveTo(env, fs);
87+
88+
Console.WriteLine("The model is saved to {0}", ModelPath);
89+
}
90+
91+
private static void PredictWithModelLoadedFromFile(IrisData sampleData)
92+
{
93+
// Test with Loaded Model from .zip file
94+
95+
using (var env = new LocalEnvironment())
96+
{
97+
ITransformer loadedModel;
98+
using (var stream = new FileStream(ModelPath, FileMode.Open, FileAccess.Read, FileShare.Read))
99+
{
100+
loadedModel = TransformerChain.LoadFrom(env, stream);
101+
}
102+
103+
// Create prediction engine and make prediction.
104+
var prediction = loadedModel.MakePredictionFunction<IrisData, IrisPrediction>(env).Predict(
105+
new IrisData()
106+
{
107+
SepalLength = 3.3f,
108+
SepalWidth = 1.6f,
109+
PetalLength = 0.2f,
110+
PetalWidth = 5.1f,
111+
});
112+
113+
Console.WriteLine();
114+
Console.WriteLine($"Clusters assigned for setosa flowers:" + prediction.SelectedClusterId);
115+
}
116+
}
117+
118+
}
119+
120+
121+
122+
// Define your data structures
123+
public class IrisData
124+
{
125+
[Column("0")]
126+
public float Label;
127+
128+
[Column("1")]
129+
public float SepalLength;
130+
131+
[Column("2")]
132+
public float SepalWidth;
133+
134+
[Column("3")]
135+
public float PetalLength;
136+
137+
[Column("4")]
138+
public float PetalWidth;
139+
140+
}
141+
142+
// IrisPrediction is the result returned from prediction operations
143+
public class IrisPrediction
144+
{
145+
[ColumnName("PredictedLabel")]
146+
public uint SelectedClusterId;
147+
148+
[ColumnName("Score")]
149+
public float[] Distance;
150+
}
151+
}

‎samples/csharp/getting-started/Clustering_Iris/Clustering_Iris/READMe.md

+100
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,100 @@
1+
# Clustering Iris Data
2+
In this introductory sample, you'll see how to use [ML.NET](https://www.microsoft.com/net/learn/apps/machine-learning-and-ai/ml-dotnet) to divide iris flowers into different groups that correspond to different types of iris. In the world of machine learning, this task is known as **clustering**.
3+
4+
## Problem
5+
To demonstrate clustering API in action, we will use three types of iris flowers: setosa, versicolor, and virginica. All of them are stored in the same dataset. Even though the type of these flowers is known, we will not use it and run clustering algorithm only on flower parameters such as petal length, petal width, etc. The task is to group all flowers into three different clusters. We would expect the flowers of different types belong to different clusters.
6+
7+
The inputs of the model are following iris parameters:
8+
* petal length
9+
* petal width
10+
* sepal length
11+
* sepal width
12+
13+
## ML task - Clustering
14+
The generalized problem of **clustering** is to group a set of objects in such a way that objects in the same group are more similar to each other than to those in other groups.
15+
16+
Some other examples of clustering:
17+
* group news articles into topics: sports, politics, tech, etc.
18+
* group customers by purchase preferences.
19+
* divide a digital image into distinct regions for border detection or object recognition.
20+
21+
Clustering can look similar to multiclass classification, but the difference is that for clustering tasks we don't know the answers for the past data. So there is no "tutor"/"supervisor" that can tell if our algorithm's prediction was right or wrong. This type of ML task is called **unsupervised learning**.
22+
23+
## Solution
24+
To solve this problem, first we will build and train an ML model. Then we will use trained model for predicting a cluster for iris flowers.
25+
26+
### 1. Build model
27+
28+
Building a model includes: uploading data (`iris-full.txt` with `TextLoader`), transforming the data so it can be used effectively by an ML algorithm (with `ColumnConcatenator`), and choosing a learning algorithm (`KMeansPlusPlusClusterer`). All of those steps are stored in a `LearningPipeline`:
29+
```CSharp
30+
// LearningPipeline holds all steps of the learning process: data, transforms, learners.
31+
using (var env = new LocalEnvironment())
32+
{
33+
// Create DataReader with data schema mapped to file's columns
34+
var reader = new TextLoader(env,
35+
new TextLoader.Arguments()
36+
{
37+
Separator = "\t",
38+
HasHeader = true,
39+
Column = new[]
40+
{
41+
new TextLoader.Column("Label", DataKind.R4, 0),
42+
new TextLoader.Column("SepalLength", DataKind.R4, 1),
43+
new TextLoader.Column("SepalWidth", DataKind.R4, 2), new TextLoader.Column("SepalWidth", DataKind.R4, 2),
44+
new TextLoader.Column("PetalLength", DataKind.R4, 3),
45+
new TextLoader.Column("PetalWidth", DataKind.R4, 4),
46+
47+
}
48+
});
49+
//Load training data
50+
IDataView trainingDataView = reader.Read(new MultiFileSource(DataPath));
51+
52+
}
53+
```
54+
### 2. Train model
55+
Training the model is a process of running the chosen algorithm on the given data. It is implemented in the `Train()` API. To perform training we just call the method and provide our data object `IrisData` and prediction object `ClusterPrediction`.
56+
```CSharp
57+
// Transform your data and add a learner
58+
// Add a learning algorithm to the pipeline. e.g.(What are characteristics of iris is this?)
59+
// Convert the Label back into original text (after converting to number in step 3)
60+
var pipeline = new ConcatEstimator(env, "Features", "SepalLength", "SepalWidth", "PetalLength", "PetalWidth")
61+
.Append(new KMeansPlusPlusTrainer(env, "Features",clustersCount:3));
62+
63+
// Create and train the model
64+
Console.WriteLine("=============== Create and Train the Model ===============");
65+
66+
var model = pipeline.Fit(trainingDataView);
67+
68+
```
69+
### 3. Consume model
70+
After the model is build and trained, we can use the `Predict()` API to predict the cluster for an iris flower and calculate the distance from given flower parameters to each cluster (each centroid of a cluster).
71+
72+
```CSharp
73+
// Test with one sample text
74+
var sampleIrisData = new IrisData()
75+
{
76+
SepalLength = 3.3f,
77+
SepalWidth = 1.6f,
78+
PetalLength = 0.2f,
79+
PetalWidth = 5.1f,
80+
};
81+
82+
var prediction = model.MakePredictionFunction<IrisData, IrisPrediction>(env).Predict(
83+
sampleIrisData);
84+
85+
Console.WriteLine($"Clusters assigned for setosa flowers:"+prediction.SelectedClusterId);
86+
```
87+
Where `TestIrisData.Setosa1` stores the information about a setosa iris flower.
88+
```CSharp
89+
internal class TestIrisData
90+
{
91+
internal static readonly IrisData Setosa1 = new IrisData()
92+
{
93+
SepalLength = 3.3f,
94+
SepalWidth = 1.6f,
95+
PetalLength = 0.2f,
96+
PetalWidth = 5.1f,
97+
};
98+
(...)
99+
}
100+
```

0 commit comments

Comments
 (0)