update docs and prepare for 0.5.0 release

README.md

@@ -35,17 +35,17 @@ Artan requires Scala 2.12, Spark 3.0+ and Python 3,6+
 This project has been published to the Maven Central Repository. When submitting jobs on your cluster, you can use
 `spark-submit` with `--packages` parameter to download all required dependencies including python packages.
 
-    spark-submit --packages='com.github.ozancicek:artan_2.12:0.4.2'
+    spark-submit --packages='com.github.ozancicek:artan_2.12:0.5.0'
 
 For SBT:
 
-    libraryDependencies += "com.github.ozancicek" %% "artan" % "0.4.2"
+    libraryDependencies += "com.github.ozancicek" %% "artan" % "0.5.0"
 
 For python:
 
     pip install artan
 
-Note that pip will only install the python dependencies. To submit pyspark jobs, `--packages='com.github.ozancicek:artan_2.12:0.4.2'` argument should be specified in order to download necessary jars.
+Note that pip will only install the python dependencies. To submit pyspark jobs, `--packages='com.github.ozancicek:artan_2.12:0.5.0'` argument should be specified in order to download necessary jars.
 
 
 ## Docs and Examples

build.sbt

@@ -20,7 +20,7 @@ val longDesc = """Online latent state estimation with Apache Spark.
 
 lazy val settings = Seq(
   scalaVersion := scalaVer,
-  version := "0.4.2",
+  version := "0.5.0",
   organization := "com.github.ozancicek",
   organizationName := "ozancicek",
   sparkVersion := sparkVer,

docs/conf.py

@@ -21,7 +21,7 @@
 author = 'Ozan Cicekci'
 
 # The full version, including alpha/beta/rc tags
-release = '0.4.2'
+release = '0.5.0'
 rst_epilog = f"""
 .. |artan_version| replace:: {release}
 """

docs/gmmguide.rst

@@ -79,6 +79,7 @@ consecutive numbers with timestamps. These consecutive numbers are binned to sim
 
         // Set initial values and hyperparams.
         val gmm = new MultivariateGaussianMixture()
+          .setMixtureCount(3)
           .setInitialWeights(Array(0.33, 0.33, 0.33))
           .setStateKeyCol("stateKey")
           .setInitialMeans(Array(Array(3.0, 5.0), Array(6.0, 6.0), Array(7.0, 1.0)))
@@ -207,7 +208,7 @@ consecutive numbers with timestamps. These consecutive numbers are binned to sim
         eye = [1.0, 0.0, 0.0, 1.0]
         gmm = MultivariateGaussianMixture()\
             .setMixtureCount(3)\
-            .setInitialWeights([0.0, 0.0, 0.0])\
+            .setInitialWeights([0.33, 0.33, 0.33])\
             .setStateKeyCol("stateKey")\
             .setInitialMeans([[3.0, 5.0], [6.0, 6.0], [7.0, 1.0]])\
             .setInitialCovariances([eye, eye, eye])\

docs/rlsguide.rst

@@ -75,6 +75,7 @@ identifying different models and their incremented index.
 
         val filter = new RecursiveLeastSquaresFilter()
           .setStateKeyCol("stateKey")
+          .setFeatureSize(3)
           .setInitialEstimate(new DenseVector(Array(0.0, 0.0, 0.0)))
           .setRegularizationMatrixFactor(10E6)
           .setForgettingFactor(0.99)
@@ -185,6 +186,7 @@ identifying different models and their incremented index.
 
         rls = RecursiveLeastSquaresFilter()\
             .setStateKeyCol("stateKey")\
+            .setFeatureSize(3)\
             .setInitialEstimate(Vectors.dense([0.0, 0.0, 0.0]))\
             .setRegularizationMatrixFactor(10E6)\
             .setForgettingFactor(0.99)

examples/src/main/python/streaming/gmm_rate_source.py

@@ -70,7 +70,7 @@
     initial_covs = [eye, eye, eye]
     gmm = MultivariateGaussianMixture()\
         .setMixtureCount(3)\
-        .setInitialWeights([0.0, 0.0, 0.0])\
+        .setInitialWeights([0.33, 0.33, 0.33])\
         .setStateKeyCol("stateKey")\
         .setInitialMeans(initial_means)\
         .setInitialCovariances(initial_covs)\

examples/src/main/python/streaming/rls_rate_source_ols.py

@@ -89,6 +89,7 @@
 
     rls = RecursiveLeastSquaresFilter()\
         .setStateKeyCol("stateKey")\
+        .setFeatureSize(3)\
         .setInitialEstimate(Vectors.dense([0.0, 0.0, 0.0]))\
         .setRegularizationMatrixFactor(10E6)\
         .setForgettingFactor(0.99)

examples/src/main/scala/com/github/ozancicek/artan/examples/streaming/GMMRateSource.scala

@@ -61,6 +61,7 @@ object GMMRateSource {
     val mixture = when(weight < 0.2, dist1).when(weight < 0.5, dist2).otherwise(dist3)
 
     val gmm = new MultivariateGaussianMixture()
+      .setMixtureCount(3)
       .setInitialWeights(Array(0.33, 0.33, 0.33))
       .setStateKeyCol("stateKey")
       .setInitialMeans(Array(Array(3.0, 5.0), Array(6.0, 6.0), Array(7.0, 1.0)))

examples/src/main/scala/com/github/ozancicek/artan/examples/streaming/RLSRateSourceOLS.scala

@@ -67,6 +67,7 @@ object RLSRateSourceOLS {
 
     val filter = new RecursiveLeastSquaresFilter()
       .setStateKeyCol("stateKey")
+      .setFeatureSize(3)
       .setInitialEstimate(new DenseVector(Array(0.0, 0.0, 0.0)))
       .setRegularizationMatrixFactor(10E6)
       .setForgettingFactor(0.99)

python/artan/mixture/mixture_params.py

@@ -343,6 +343,15 @@ class MixtureParams(HasSampleCol, HasStepSize, HasStepSizeCol, HasInitialWeights
                     HasMinibatchSizeCol, HasUpdateHoldoutCol, HasBatchTrainEnabled, HasBatchTrainMaxIter,
                     HasBatchTrainTol, HasMixtureCount):
 
+    def setMixtureCount(self, value):
+        """
+        Sets the number of components in the finite mixture
+
+        :param value: Int
+        :return: MixtureTransformer
+        """
+        return self._set(mixtureCount=value)
+
     def setSampleCol(self, value):
         """
         Sets the sample column for the mixture model inputs. Depending on the mixture distribution, sample type should

setup.py

@@ -20,7 +20,7 @@
 from setuptools import setup
 
 
-VERSION = "0.4.2"
+VERSION = "0.5.0"
 
 with open("README.md") as f:
     long_description = f.read()