Github user tillrohrmann commented on a diff in the pull request:
https://github.com/apache/flink/pull/792#discussion_r31896996
--- Diff: docs/libs/ml/quickstart.md ---
@@ -24,4 +24,198 @@ under the License.
* This will be replaced by the TOC
{:toc}
-Coming soon.
+## Introduction
+
+FlinkML is designed to make learning from your data a straight-forward
process, abstracting away
+the complexities that usually come with having to deal with big data
learning tasks. In this
+quick-start guide we will show just how easy it is to solve a simple
supervised learning problem
+using FlinkML. But first some basics, feel free to skip the next few lines
if you're already
+familiar with Machine Learning (ML)
+
+As defined by Murphy [cite ML-APP] ML deals with detecting patterns in
data, and using those
+learned patterns to make predictions about the future. We can categorize
most ML algorithms into
+two major categories: Supervised and Unsupervised Learning.
+
+* Supervised Learning deals with learning a function (mapping) from a set
of inputs
+(predictors) to a set of outputs. The learning is done using a __training
set__ of (input,
+output) pairs that we use to approximate the mapping function. Supervised
learning problems are
+further divided into classification and regression problems. In
classification problems we try to
+predict the __class__ that an example belongs to, for example whether a
user is going to click on
+an ad or not. Regression problems are about predicting (real) numerical
values, often called the dependent
+variable, for example what the temperature will be tomorrow.
+
+* Unsupervised learning deals with discovering patterns and regularities
in the data. An example
+of this would be __clustering__, where we try to discover groupings of the
data from the
+descriptive features. Unsupervised learning can also be used for feature
selection, for example
+through [principal components
analysis](https://en.wikipedia.org/wiki/Principal_component_analysis).
+
+## Loading data
+
+For loading data to be used with FlinkML we can use the ETL capabilities
of Flink, or specialized
+functions for formatted data, such as the LibSVM format. For supervised
learning problems it is
+common to use the `LabeledVector` class to represent the `(features,
label)` examples. A `LabeledVector`
+object will have a FlinkML `Vector` member representing the features of
the example and a `Double`
+member which represents the label, which could be the class in a
classification problem, or the dependent
+variable for a regression problem.
+
+# TODO: Get dataset that has separate train and test sets
+As an example, we can use the Breast Cancer Wisconsin (Diagnostic) Data
Set, which you can
+[download from the UCI ML
repository](http://archive.ics.uci.edu/ml/machine-learning-databases/breast-cancer-wisconsin/breast-cancer-wisconsin.data).
+
+We can load the data as a `DataSet[String]` first:
+
+{% highlight scala %}
+
+val cancer = env.readCsvFile[(String, String, String, String, String,
String, String, String, String, String,
String)]("/path/to/breast-cancer-wisconsin.data")
+
+{% endhighlight %}
+
+The dataset has some missing values indicated by `?`. We can filter those
rows out and
+then transform the data into a `DataSet[LabeledVector]`. This will allow
us to use the
+dataset with the FlinkML classification algorithms.
+
+{% highlight scala %}
+
+val cancerLV = cancer
+ .map(_.productIterator.toList)
+ .filter(!_.contains("?"))
+ .map{list =>
+ val numList = list.map(_.asInstanceOf[String].toDouble)
+ LabeledVector(numList(11), DenseVector(numList.take(10).toArray))
+ }
+
+{% endhighlight %}
+
+We can then use this data to train a learner.
+
+A common format for ML datasets is the LibSVM format and a number of
datasets using that format can be
+found [in the LibSVM datasets
website](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/). FlinkML
provides utilities for loading
+datasets using the LibSVM format through the `readLibSVM` function
available through the MLUtils object.
+You can also save datasets in the LibSVM format using the `writeLibSVM`
function.
+Let's import the Adult (a9a) dataset. You can download the
+[training set
here](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/a8a)
+and the [test set
here](http://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/a8a.t).
+
+We can simply import the dataset then using:
+
+{% highlight scala %}
+
+val adultTrain = MLUtils.readLibSVM("path/to/a8a")
+val adultTest = MLUtils.readLibSVM("path/to/a8a.t")
+
+{% endhighlight %}
+
+This gives us a `DataSet[LabeledVector]` that we will use in the following
section to create a classifier.
+
+Due to an error in the test dataset we have to adjust the test data using
the following code, to
+ensure that the dimensionality of all test examples is 123, as with the
training set:
+
+{% highlight scala %}
+
+val adjustedTest = adultTest.map{lv =>
+ val vec = lv.vector.asBreeze
+ val padded = vec.padTo(123, 0.0).toDenseVector
+ val fvec = padded.fromBreeze
+ LabeledVector(lv.label, fvec)
+ }
+
+{% endhighlight %}
+
+## Classification
+
+Once we have imported the dataset we can train a `Predictor` such as a
linear SVM classifier.
+
+{% highlight scala %}
+
+val svm = SVM()
+ .setBlocks(env.getParallelism)
+ .setIterations(100)
+ .setRegularization(0.001)
+ .setStepsize(0.1)
+ .setSeed(42)
+
+svm.fit(adultTrain)
+
+{% endhighlight %}
+
+Let's now make predictions on the test set and see how well we do in terms
of absolute error
+We will also create a function that thresholds the predictions to the {-1,
1} scale that the
+dataset uses.
+
+{% highlight scala %}
+
+def thresholdPredictions(predictions: DataSet[(Double, Double)])
+: DataSet[(Double, Double)] = {
+ predictions.map {
+ truthPrediction =>
+ val truth = truthPrediction._1
+ val prediction = truthPrediction._2
+ val thresholdedPrediction = if (prediction > 0.0) 1.0 else -1.0
+ (truth, thresholdedPrediction)
+ }
+}
+
+val predictionPairs = thresholdPredictions(svm.predict(adjustedTest))
+
+val absoluteErrorSum = predictionPairs.collect().map{
+ case (truth, prediction) => Math.abs(truth - prediction)}.sum
+
+println(s"Absolute error: $absoluteErrorSum")
+
+{% endhighlight %}
+
+Next we will see if we can improve the performance by pre-processing our
data.
+
+## Data pre-processing and pipelines
+
+A pre-processing step that is often encouraged when using SVM
classification is scaling
+the input features to the [0, 1] range, in order to avoid features with
extreme values dominating the rest.
+FlinkML has a number of `Transformers` such as `StandardScaler` that are
used to pre-process data, and a key feature is the ability to
+chain `Transformers` and `Predictors` together. This allows us to run the
same pipeline of transformations and make predictions
+on the train and test data in a straight-forward and type-safe manner. You
can read more on the pipeline system of FlinkML,
+[here](pipelines.html).
+
+Let first create a scaling transformer for the features in our dataset,
and chain it to a new SVM classifier.
+
+{% highlight scala %}
+
+import org.apache.flink.ml.preprocessing.StandardScaler
+
+val scaler = StandardScaler()
+scaler.fit(adultTrain)
+
+val scaledSVM = scaler.chainPredictor(svm)
+
+{% endhighlight %}
+
+We can now use our newly created pipeline to make predictions on the test
set.
+First we call fit again, to train the scaler and the SVM classifier.
+The data of the test set will then be automatically scaled before being
passed on to the SVM to
+make predictions.
+
+{% highlight scala %}
+
+scaledSVM.fit(adultTrain)
+
+val predictionPairsScaled=
thresholdPredictions(scaledSVM.predict(predictionsScaled))
+
+val absoluteErrorSumScaled = predictionPairs.collect().map{
+ case (truth, prediction) => Math.abs(truth - prediction)}.sum
+
+println(s"Absolute error with scaled features: $absoluteErrorSumScaled")
+
+{% endhighlight %}
+
+The effect that the transformation has on the rror for this dataset is a
bit unpredictable.
+In reality the scaling transformation does
+not fit the dataset we are using, since the features are translated
categorical features and as
+such, operations like normalization and standard scaling do not make much
sense.
--- End diff --
Hmm, maybe we should try to find a data set where scaling actually helps
us. Otherwise users might ask themselves, why using this feature at all.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
