(DEPRECATED) Apache Flink User Mailing List archive.

SVM Multiclass classification

Classic

List

Threaded

2 messages Options

Kürşat Kurt

SVM Multiclass classification

Hi;

I am trying to classify documents.

When i try to predict (same of training set) there is only 1 and -1 predictions.

Accuracy is 0%.

Can you help me please?

val env = ExecutionEnvironment.getExecutionEnvironment

val training = Seq(

new LabeledVector(1.0, new SparseVector(10, Array(0, 2, 3), Array(1.0, 1.0, 1.0))),

new LabeledVector(1.0, new SparseVector(10, Array(0, 1, 5, 9), Array(1.0, 1.0, 1.0, 1.0))),

new LabeledVector(0.0, new SparseVector(10, Array(0, 2), Array(0.0, 1.0))),

new LabeledVector(0.0, new SparseVector(10, Array(0), Array(0.0))),

new LabeledVector(2.0, new SparseVector(10, Array(0, 2), Array(0.0, 1.0))),

new LabeledVector(2.0, new SparseVector(10, Array(0), Array(0.0))),

new LabeledVector(1.0, new SparseVector(10, Array(0, 3), Array(1.0, 1.0))),

new LabeledVector(0.0, new SparseVector(10, Array(0, 2, 3), Array(0.0, 1.0, 1.0))),

new LabeledVector(2.0, new SparseVector(10, Array(0, 7, 9), Array(0.0, 1.0))),

new LabeledVector(2.0, new SparseVector(10, Array(2,3,4), Array(0.0,1.0,1.0))),

new LabeledVector(2.0, new SparseVector(10, Array(0, 3), Array(1.0, 1.0))),

new LabeledVector(0.0, new SparseVector(10, Array(2, 3,9), Array(1.0, 0.0, 1.0)))

);

val trainingDS = env.fromCollection(training)

val testingDS = env.fromCollection(training)

val svm = new SVM().setBlocks(env.getParallelism)

svm.fit(trainingDS)

val predictions = svm.evaluate(testingDS.map(x => (x.vector, x.label)))

predictions.print();

Sample output:

(1.0,1.0)

(0.0,1.0)

(0.0,-1.0)

(2.0,1.0)

(2.0,-1.0)

(1.0,1.0)

(0.0,1.0)

(2.0,1.0)

(0.0,1.0)

Theodore Vasiloudis

Re: SVM Multiclass classification

Hello Kursat,

As noted in the documentation, the SVM implementation is for binary classification only for the time being.

Regards,
Theodore

--
Sent from a mobile device. May contain autocorrect errors.

On Oct 13, 2016 8:53 PM, "Kürşat Kurt" <[hidden email]> wrote:

Hi;

I am trying to classify documents.
When i try to predict (same of training set) there is only 1 and -1 predictions.
Accuracy is 0%.

Can you help me please?

val env = ExecutionEnvironment.getExecutionEnvironment
    val training = Seq(
      new LabeledVector(1.0, new SparseVector(10, Array(0, 2, 3), Array(1.0, 1.0, 1.0))),
      new LabeledVector(1.0, new SparseVector(10, Array(0, 1, 5, 9), Array(1.0, 1.0, 1.0, 1.0))),
      new LabeledVector(0.0, new SparseVector(10, Array(0, 2), Array(0.0, 1.0))),
      new LabeledVector(0.0, new SparseVector(10, Array(0), Array(0.0))),
      new LabeledVector(2.0, new SparseVector(10, Array(0, 2), Array(0.0, 1.0))),
      new LabeledVector(2.0, new SparseVector(10, Array(0), Array(0.0))),
      new LabeledVector(1.0, new SparseVector(10, Array(0, 3), Array(1.0, 1.0))),
      new LabeledVector(0.0, new SparseVector(10, Array(0, 2, 3), Array(0.0, 1.0, 1.0))),
      new LabeledVector(2.0, new SparseVector(10, Array(0, 7, 9), Array(0.0, 1.0))),
      new LabeledVector(2.0, new SparseVector(10, Array(2,3,4), Array(0.0,1.0,1.0))),
      new LabeledVector(2.0, new SparseVector(10, Array(0, 3), Array(1.0, 1.0))),
      new LabeledVector(0.0, new SparseVector(10, Array(2, 3,9), Array(1.0, 0.0, 1.0)))

    );
    val trainingDS = env.fromCollection(training)
    val testingDS = env.fromCollection(training)
    val svm = new SVM().setBlocks(env.getParallelism)
    svm.fit(trainingDS)
    val predictions = svm.evaluate(testingDS.map(x => (x.vector, x.label)))
    predictions.print();

Sample output:

(1.0,1.0)
(1.0,1.0)
(0.0,1.0)
(0.0,-1.0)
(2.0,1.0)
(2.0,-1.0)
(1.0,1.0)
(0.0,1.0)
(2.0,1.0)
(2.0,1.0)
(2.0,1.0)
(0.0,1.0)