You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@spark.apache.org by me...@apache.org on 2015/02/20 11:31:36 UTC
[1/2] spark git commit: [SPARK-5867] [SPARK-5892] [doc] [ml] [mllib]
Doc cleanups for 1.3 release
Repository: spark
Updated Branches:
refs/heads/master d3dfebebc -> 4a17eedb1
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleTextClassificationPipeline.java
----------------------------------------------------------------------
diff --git a/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleTextClassificationPipeline.java b/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleTextClassificationPipeline.java
index fdcfc88..ef1ec10 100644
--- a/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleTextClassificationPipeline.java
+++ b/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleTextClassificationPipeline.java
@@ -30,8 +30,8 @@ import org.apache.spark.ml.classification.LogisticRegression;
import org.apache.spark.ml.feature.HashingTF;
import org.apache.spark.ml.feature.Tokenizer;
import org.apache.spark.sql.DataFrame;
-import org.apache.spark.sql.SQLContext;
import org.apache.spark.sql.Row;
+import org.apache.spark.sql.SQLContext;
/**
* A simple text classification pipeline that recognizes "spark" from input text. It uses the Java
@@ -82,9 +82,8 @@ public class JavaSimpleTextClassificationPipeline {
DataFrame test = jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
// Make predictions on test documents.
- model.transform(test).registerTempTable("prediction");
- DataFrame predictions = jsql.sql("SELECT id, text, score, prediction FROM prediction");
- for (Row r: predictions.collect()) {
+ DataFrame predictions = model.transform(test);
+ for (Row r: predictions.select("id", "text", "probability", "prediction").collect()) {
System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/python/ml/simple_text_classification_pipeline.py
----------------------------------------------------------------------
diff --git a/examples/src/main/python/ml/simple_text_classification_pipeline.py b/examples/src/main/python/ml/simple_text_classification_pipeline.py
index b4d9355..d281f4f 100644
--- a/examples/src/main/python/ml/simple_text_classification_pipeline.py
+++ b/examples/src/main/python/ml/simple_text_classification_pipeline.py
@@ -16,10 +16,10 @@
#
from pyspark import SparkContext
-from pyspark.sql import SQLContext, Row
from pyspark.ml import Pipeline
-from pyspark.ml.feature import HashingTF, Tokenizer
from pyspark.ml.classification import LogisticRegression
+from pyspark.ml.feature import HashingTF, Tokenizer
+from pyspark.sql import Row, SQLContext
"""
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/scala/org/apache/spark/examples/ml/CrossValidatorExample.scala
----------------------------------------------------------------------
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/CrossValidatorExample.scala b/examples/src/main/scala/org/apache/spark/examples/ml/CrossValidatorExample.scala
index 7ab892c..6c0af20 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/CrossValidatorExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/CrossValidatorExample.scala
@@ -100,7 +100,7 @@ object CrossValidatorExample {
Document(7L, "apache hadoop")))
// Make predictions on test documents. cvModel uses the best model found (lrModel).
- cvModel.transform(test.toDF)
+ cvModel.transform(test.toDF())
.select("id", "text", "probability", "prediction")
.collect()
.foreach { case Row(id: Long, text: String, prob: Vector, prediction: Double) =>
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/scala/org/apache/spark/examples/ml/MovieLensALS.scala
----------------------------------------------------------------------
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/MovieLensALS.scala b/examples/src/main/scala/org/apache/spark/examples/ml/MovieLensALS.scala
index 96b2dd4..25f2111 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/MovieLensALS.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/MovieLensALS.scala
@@ -93,8 +93,8 @@ object MovieLensALS {
| bin/spark-submit --class org.apache.spark.examples.ml.MovieLensALS \
| examples/target/scala-*/spark-examples-*.jar \
| --rank 10 --maxIter 15 --regParam 0.1 \
- | --movies path/to/movielens/movies.dat \
- | --ratings path/to/movielens/ratings.dat
+ | --movies data/mllib/als/sample_movielens_movies.txt \
+ | --ratings data/mllib/als/sample_movielens_ratings.txt
""".stripMargin)
}
@@ -157,17 +157,23 @@ object MovieLensALS {
println(s"Test RMSE = $rmse.")
// Inspect false positives.
- predictions.registerTempTable("prediction")
- sc.textFile(params.movies).map(Movie.parseMovie).toDF().registerTempTable("movie")
- sqlContext.sql(
- """
- |SELECT userId, prediction.movieId, title, rating, prediction
- | FROM prediction JOIN movie ON prediction.movieId = movie.movieId
- | WHERE rating <= 1 AND prediction >= 4
- | LIMIT 100
- """.stripMargin)
- .collect()
- .foreach(println)
+ // Note: We reference columns in 2 ways:
+ // (1) predictions("movieId") lets us specify the movieId column in the predictions
+ // DataFrame, rather than the movieId column in the movies DataFrame.
+ // (2) $"userId" specifies the userId column in the predictions DataFrame.
+ // We could also write predictions("userId") but do not have to since
+ // the movies DataFrame does not have a column "userId."
+ val movies = sc.textFile(params.movies).map(Movie.parseMovie).toDF()
+ val falsePositives = predictions.join(movies)
+ .where((predictions("movieId") === movies("movieId"))
+ && ($"rating" <= 1) && ($"prediction" >= 4))
+ .select($"userId", predictions("movieId"), $"title", $"rating", $"prediction")
+ val numFalsePositives = falsePositives.count()
+ println(s"Found $numFalsePositives false positives")
+ if (numFalsePositives > 0) {
+ println(s"Example false positives:")
+ falsePositives.limit(100).collect().foreach(println)
+ }
sc.stop()
}
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/scala/org/apache/spark/examples/ml/SimpleParamsExample.scala
----------------------------------------------------------------------
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/SimpleParamsExample.scala b/examples/src/main/scala/org/apache/spark/examples/ml/SimpleParamsExample.scala
index e8af5c1..bf80514 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/SimpleParamsExample.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/SimpleParamsExample.scala
@@ -40,8 +40,8 @@ object SimpleParamsExample {
import sqlContext.implicits._
// Prepare training data.
- // We use LabeledPoint, which is a case class. Spark SQL can convert RDDs of Java Beans
- // into DataFrames, where it uses the bean metadata to infer the schema.
+ // We use LabeledPoint, which is a case class. Spark SQL can convert RDDs of case classes
+ // into DataFrames, where it uses the case class metadata to infer the schema.
val training = sc.parallelize(Seq(
LabeledPoint(1.0, Vectors.dense(0.0, 1.1, 0.1)),
LabeledPoint(0.0, Vectors.dense(2.0, 1.0, -1.0)),
@@ -94,7 +94,7 @@ object SimpleParamsExample {
.select("features", "label", "myProbability", "prediction")
.collect()
.foreach { case Row(features: Vector, label: Double, prob: Vector, prediction: Double) =>
- println("($features, $label) -> prob=$prob, prediction=$prediction")
+ println(s"($features, $label) -> prob=$prob, prediction=$prediction")
}
sc.stop()
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/scala/org/apache/spark/examples/ml/SimpleTextClassificationPipeline.scala
----------------------------------------------------------------------
diff --git a/examples/src/main/scala/org/apache/spark/examples/ml/SimpleTextClassificationPipeline.scala b/examples/src/main/scala/org/apache/spark/examples/ml/SimpleTextClassificationPipeline.scala
index a11db6f..6772efd 100644
--- a/examples/src/main/scala/org/apache/spark/examples/ml/SimpleTextClassificationPipeline.scala
+++ b/examples/src/main/scala/org/apache/spark/examples/ml/SimpleTextClassificationPipeline.scala
@@ -83,7 +83,7 @@ object SimpleTextClassificationPipeline {
.select("id", "text", "probability", "prediction")
.collect()
.foreach { case Row(id: Long, text: String, prob: Vector, prediction: Double) =>
- println("($id, $text) --> prob=$prob, prediction=$prediction")
+ println(s"($id, $text) --> prob=$prob, prediction=$prediction")
}
sc.stop()
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala
----------------------------------------------------------------------
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala b/mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala
index 80584ef..568b653 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/clustering/GaussianMixture.scala
@@ -19,12 +19,10 @@ package org.apache.spark.mllib.clustering
import scala.collection.mutable.IndexedSeq
-import breeze.linalg.{diag, DenseMatrix => BreezeMatrix, DenseVector => BDV, SparseVector => BSV,
- Transpose, Vector => BV}
+import breeze.linalg.{diag, DenseMatrix => BreezeMatrix, DenseVector => BDV, Vector => BV}
import org.apache.spark.annotation.Experimental
-import org.apache.spark.mllib.linalg.{BLAS, DenseVector, DenseMatrix, Matrices,
- SparseVector, Vector, Vectors}
+import org.apache.spark.mllib.linalg.{BLAS, DenseMatrix, Matrices, Vector, Vectors}
import org.apache.spark.mllib.stat.distribution.MultivariateGaussian
import org.apache.spark.mllib.util.MLUtils
import org.apache.spark.rdd.RDD
@@ -43,7 +41,11 @@ import org.apache.spark.util.Utils
* less than convergenceTol, or until it has reached the max number of iterations.
* While this process is generally guaranteed to converge, it is not guaranteed
* to find a global optimum.
- *
+ *
+ * Note: For high-dimensional data (with many features), this algorithm may perform poorly.
+ * This is due to high-dimensional data (a) making it difficult to cluster at all (based
+ * on statistical/theoretical arguments) and (b) numerical issues with Gaussian distributions.
+ *
* @param k The number of independent Gaussians in the mixture model
* @param convergenceTol The maximum change in log-likelihood at which convergence
* is considered to have occurred.
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/mllib/src/main/scala/org/apache/spark/mllib/linalg/Matrices.scala
----------------------------------------------------------------------
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/linalg/Matrices.scala b/mllib/src/main/scala/org/apache/spark/mllib/linalg/Matrices.scala
index 89b3867..0e4a4d0 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/linalg/Matrices.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/linalg/Matrices.scala
@@ -706,7 +706,7 @@ object Matrices {
}
/**
- * Generate a `DenseMatrix` consisting of zeros.
+ * Generate a `Matrix` consisting of zeros.
* @param numRows number of rows of the matrix
* @param numCols number of columns of the matrix
* @return `Matrix` with size `numRows` x `numCols` and values of zeros
@@ -778,8 +778,8 @@ object Matrices {
SparseMatrix.sprandn(numRows, numCols, density, rng)
/**
- * Generate a diagonal matrix in `DenseMatrix` format from the supplied values.
- * @param vector a `Vector` tat will form the values on the diagonal of the matrix
+ * Generate a diagonal matrix in `Matrix` format from the supplied values.
+ * @param vector a `Vector` that will form the values on the diagonal of the matrix
* @return Square `Matrix` with size `values.length` x `values.length` and `values`
* on the diagonal
*/
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala
----------------------------------------------------------------------
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala b/mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala
index 480bbfb..4bdcb28 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/linalg/Vectors.scala
@@ -247,7 +247,7 @@ object Vectors {
}
/**
- * Creates a dense vector of all zeros.
+ * Creates a vector of all zeros.
*
* @param size vector size
* @return a zero vector
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/mllib/src/main/scala/org/apache/spark/mllib/optimization/Gradient.scala
----------------------------------------------------------------------
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/optimization/Gradient.scala b/mllib/src/main/scala/org/apache/spark/mllib/optimization/Gradient.scala
index 0acdab7..8bfa0d2 100644
--- a/mllib/src/main/scala/org/apache/spark/mllib/optimization/Gradient.scala
+++ b/mllib/src/main/scala/org/apache/spark/mllib/optimization/Gradient.scala
@@ -63,10 +63,12 @@ abstract class Gradient extends Serializable {
* http://statweb.stanford.edu/~tibs/ElemStatLearn/ , Eq. (4.17) on page 119 gives the formula of
* multinomial logistic regression model. A simple calculation shows that
*
+ * {{{
* P(y=0|x, w) = 1 / (1 + \sum_i^{K-1} \exp(x w_i))
* P(y=1|x, w) = exp(x w_1) / (1 + \sum_i^{K-1} \exp(x w_i))
* ...
* P(y=K-1|x, w) = exp(x w_{K-1}) / (1 + \sum_i^{K-1} \exp(x w_i))
+ * }}}
*
* for K classes multiclass classification problem.
*
@@ -75,9 +77,11 @@ abstract class Gradient extends Serializable {
* will be (K-1) * N.
*
* As a result, the loss of objective function for a single instance of data can be written as
+ * {{{
* l(w, x) = -log P(y|x, w) = -\alpha(y) log P(y=0|x, w) - (1-\alpha(y)) log P(y|x, w)
* = log(1 + \sum_i^{K-1}\exp(x w_i)) - (1-\alpha(y)) x w_{y-1}
* = log(1 + \sum_i^{K-1}\exp(margins_i)) - (1-\alpha(y)) margins_{y-1}
+ * }}}
*
* where \alpha(i) = 1 if i != 0, and
* \alpha(i) = 0 if i == 0,
@@ -86,14 +90,16 @@ abstract class Gradient extends Serializable {
* For optimization, we have to calculate the first derivative of the loss function, and
* a simple calculation shows that
*
+ * {{{
* \frac{\partial l(w, x)}{\partial w_{ij}}
* = (\exp(x w_i) / (1 + \sum_k^{K-1} \exp(x w_k)) - (1-\alpha(y)\delta_{y, i+1})) * x_j
* = multiplier_i * x_j
+ * }}}
*
* where \delta_{i, j} = 1 if i == j,
* \delta_{i, j} = 0 if i != j, and
- * multiplier
- * = \exp(margins_i) / (1 + \sum_k^{K-1} \exp(margins_i)) - (1-\alpha(y)\delta_{y, i+1})
+ * multiplier =
+ * \exp(margins_i) / (1 + \sum_k^{K-1} \exp(margins_i)) - (1-\alpha(y)\delta_{y, i+1})
*
* If any of margins is larger than 709.78, the numerical computation of multiplier and loss
* function will be suffered from arithmetic overflow. This issue occurs when there are outliers
@@ -103,10 +109,12 @@ abstract class Gradient extends Serializable {
* Fortunately, when max(margins) = maxMargin > 0, the loss function and the multiplier can be
* easily rewritten into the following equivalent numerically stable formula.
*
+ * {{{
* l(w, x) = log(1 + \sum_i^{K-1}\exp(margins_i)) - (1-\alpha(y)) margins_{y-1}
* = log(\exp(-maxMargin) + \sum_i^{K-1}\exp(margins_i - maxMargin)) + maxMargin
* - (1-\alpha(y)) margins_{y-1}
* = log(1 + sum) + maxMargin - (1-\alpha(y)) margins_{y-1}
+ * }}}
*
* where sum = \exp(-maxMargin) + \sum_i^{K-1}\exp(margins_i - maxMargin) - 1.
*
@@ -115,8 +123,10 @@ abstract class Gradient extends Serializable {
*
* For multiplier, similar trick can be applied as the following,
*
+ * {{{
* multiplier = \exp(margins_i) / (1 + \sum_k^{K-1} \exp(margins_i)) - (1-\alpha(y)\delta_{y, i+1})
* = \exp(margins_i - maxMargin) / (1 + sum) - (1-\alpha(y)\delta_{y, i+1})
+ * }}}
*
* where each term in \exp is also smaller than zero, so overflow is not a concern.
*
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/docs/pyspark.mllib.rst
----------------------------------------------------------------------
diff --git a/python/docs/pyspark.mllib.rst b/python/docs/pyspark.mllib.rst
index 21f66ca..b706c5e 100644
--- a/python/docs/pyspark.mllib.rst
+++ b/python/docs/pyspark.mllib.rst
@@ -7,7 +7,7 @@ pyspark.mllib.classification module
.. automodule:: pyspark.mllib.classification
:members:
:undoc-members:
- :show-inheritance:
+ :inherited-members:
pyspark.mllib.clustering module
-------------------------------
@@ -15,7 +15,6 @@ pyspark.mllib.clustering module
.. automodule:: pyspark.mllib.clustering
:members:
:undoc-members:
- :show-inheritance:
pyspark.mllib.feature module
-------------------------------
@@ -39,7 +38,6 @@ pyspark.mllib.random module
.. automodule:: pyspark.mllib.random
:members:
:undoc-members:
- :show-inheritance:
pyspark.mllib.recommendation module
-----------------------------------
@@ -47,7 +45,6 @@ pyspark.mllib.recommendation module
.. automodule:: pyspark.mllib.recommendation
:members:
:undoc-members:
- :show-inheritance:
pyspark.mllib.regression module
-------------------------------
@@ -55,7 +52,7 @@ pyspark.mllib.regression module
.. automodule:: pyspark.mllib.regression
:members:
:undoc-members:
- :show-inheritance:
+ :inherited-members:
pyspark.mllib.stat module
-------------------------
@@ -63,7 +60,6 @@ pyspark.mllib.stat module
.. automodule:: pyspark.mllib.stat
:members:
:undoc-members:
- :show-inheritance:
pyspark.mllib.tree module
-------------------------
@@ -71,7 +67,7 @@ pyspark.mllib.tree module
.. automodule:: pyspark.mllib.tree
:members:
:undoc-members:
- :show-inheritance:
+ :inherited-members:
pyspark.mllib.util module
-------------------------
@@ -79,4 +75,3 @@ pyspark.mllib.util module
.. automodule:: pyspark.mllib.util
:members:
:undoc-members:
- :show-inheritance:
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/ml/classification.py
----------------------------------------------------------------------
diff --git a/python/pyspark/ml/classification.py b/python/pyspark/ml/classification.py
index b6de749..4ff7463 100644
--- a/python/pyspark/ml/classification.py
+++ b/python/pyspark/ml/classification.py
@@ -15,10 +15,11 @@
# limitations under the License.
#
-from pyspark.ml.util import inherit_doc, keyword_only
+from pyspark.ml.util import keyword_only
from pyspark.ml.wrapper import JavaEstimator, JavaModel
from pyspark.ml.param.shared import HasFeaturesCol, HasLabelCol, HasPredictionCol, HasMaxIter,\
HasRegParam
+from pyspark.mllib.common import inherit_doc
__all__ = ['LogisticRegression', 'LogisticRegressionModel']
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/ml/feature.py
----------------------------------------------------------------------
diff --git a/python/pyspark/ml/feature.py b/python/pyspark/ml/feature.py
index f1ddbb4..433b4fb 100644
--- a/python/pyspark/ml/feature.py
+++ b/python/pyspark/ml/feature.py
@@ -16,8 +16,9 @@
#
from pyspark.ml.param.shared import HasInputCol, HasOutputCol, HasNumFeatures
-from pyspark.ml.util import inherit_doc, keyword_only
+from pyspark.ml.util import keyword_only
from pyspark.ml.wrapper import JavaTransformer
+from pyspark.mllib.common import inherit_doc
__all__ = ['Tokenizer', 'HashingTF']
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/ml/pipeline.py
----------------------------------------------------------------------
diff --git a/python/pyspark/ml/pipeline.py b/python/pyspark/ml/pipeline.py
index 18d8a58..5233c58 100644
--- a/python/pyspark/ml/pipeline.py
+++ b/python/pyspark/ml/pipeline.py
@@ -18,7 +18,8 @@
from abc import ABCMeta, abstractmethod
from pyspark.ml.param import Param, Params
-from pyspark.ml.util import inherit_doc, keyword_only
+from pyspark.ml.util import keyword_only
+from pyspark.mllib.common import inherit_doc
__all__ = ['Estimator', 'Transformer', 'Pipeline', 'PipelineModel']
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/ml/util.py
----------------------------------------------------------------------
diff --git a/python/pyspark/ml/util.py b/python/pyspark/ml/util.py
index 81d3f08..6f7f39c 100644
--- a/python/pyspark/ml/util.py
+++ b/python/pyspark/ml/util.py
@@ -19,20 +19,6 @@ from functools import wraps
import uuid
-def inherit_doc(cls):
- for name, func in vars(cls).items():
- # only inherit docstring for public functions
- if name.startswith("_"):
- continue
- if not func.__doc__:
- for parent in cls.__bases__:
- parent_func = getattr(parent, name, None)
- if parent_func and getattr(parent_func, "__doc__", None):
- func.__doc__ = parent_func.__doc__
- break
- return cls
-
-
def keyword_only(func):
"""
A decorator that forces keyword arguments in the wrapped method
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/ml/wrapper.py
----------------------------------------------------------------------
diff --git a/python/pyspark/ml/wrapper.py b/python/pyspark/ml/wrapper.py
index 9e12ddc..4bae96f 100644
--- a/python/pyspark/ml/wrapper.py
+++ b/python/pyspark/ml/wrapper.py
@@ -21,7 +21,7 @@ from pyspark import SparkContext
from pyspark.sql import DataFrame
from pyspark.ml.param import Params
from pyspark.ml.pipeline import Estimator, Transformer
-from pyspark.ml.util import inherit_doc
+from pyspark.mllib.common import inherit_doc
def _jvm():
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/__init__.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/__init__.py b/python/pyspark/mllib/__init__.py
index c321762..6449800 100644
--- a/python/pyspark/mllib/__init__.py
+++ b/python/pyspark/mllib/__init__.py
@@ -19,7 +19,7 @@
Python bindings for MLlib.
"""
-# MLlib currently needs and NumPy 1.4+, so complain if lower
+# MLlib currently needs NumPy 1.4+, so complain if lower
import numpy
if numpy.version.version < '1.4':
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/clustering.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/clustering.py b/python/pyspark/mllib/clustering.py
index f6b97ab..949db57 100644
--- a/python/pyspark/mllib/clustering.py
+++ b/python/pyspark/mllib/clustering.py
@@ -152,7 +152,7 @@ class GaussianMixtureModel(object):
class GaussianMixture(object):
"""
- Estimate model parameters with the expectation-maximization algorithm.
+ Learning algorithm for Gaussian Mixtures using the expectation-maximization algorithm.
:param data: RDD of data points
:param k: Number of components
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/common.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/common.py b/python/pyspark/mllib/common.py
index 3c5ee66..621591c 100644
--- a/python/pyspark/mllib/common.py
+++ b/python/pyspark/mllib/common.py
@@ -134,3 +134,20 @@ class JavaModelWrapper(object):
def call(self, name, *a):
"""Call method of java_model"""
return callJavaFunc(self._sc, getattr(self._java_model, name), *a)
+
+
+def inherit_doc(cls):
+ """
+ A decorator that makes a class inherit documentation from its parents.
+ """
+ for name, func in vars(cls).items():
+ # only inherit docstring for public functions
+ if name.startswith("_"):
+ continue
+ if not func.__doc__:
+ for parent in cls.__bases__:
+ parent_func = getattr(parent, name, None)
+ if parent_func and getattr(parent_func, "__doc__", None):
+ func.__doc__ = parent_func.__doc__
+ break
+ return cls
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/regression.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/regression.py b/python/pyspark/mllib/regression.py
index 2100601..21751cc 100644
--- a/python/pyspark/mllib/regression.py
+++ b/python/pyspark/mllib/regression.py
@@ -18,7 +18,7 @@
import numpy as np
from numpy import array
-from pyspark.mllib.common import callMLlibFunc
+from pyspark.mllib.common import callMLlibFunc, inherit_doc
from pyspark.mllib.linalg import SparseVector, _convert_to_vector
__all__ = ['LabeledPoint', 'LinearModel', 'LinearRegressionModel', 'RidgeRegressionModel',
@@ -69,6 +69,7 @@ class LinearModel(object):
return "(weights=%s, intercept=%r)" % (self._coeff, self._intercept)
+@inherit_doc
class LinearRegressionModelBase(LinearModel):
"""A linear regression model.
@@ -89,6 +90,7 @@ class LinearRegressionModelBase(LinearModel):
return self.weights.dot(x) + self.intercept
+@inherit_doc
class LinearRegressionModel(LinearRegressionModelBase):
"""A linear regression model derived from a least-squares fit.
@@ -162,7 +164,7 @@ class LinearRegressionWithSGD(object):
@param intercept: Boolean parameter which indicates the use
or not of the augmented representation for
training data (i.e. whether bias features
- are activated or not).
+ are activated or not). (default: False)
"""
def train(rdd, i):
return callMLlibFunc("trainLinearRegressionModelWithSGD", rdd, int(iterations),
@@ -172,6 +174,7 @@ class LinearRegressionWithSGD(object):
return _regression_train_wrapper(train, LinearRegressionModel, data, initialWeights)
+@inherit_doc
class LassoModel(LinearRegressionModelBase):
"""A linear regression model derived from a least-squares fit with an
@@ -218,6 +221,7 @@ class LassoWithSGD(object):
return _regression_train_wrapper(train, LassoModel, data, initialWeights)
+@inherit_doc
class RidgeRegressionModel(LinearRegressionModelBase):
"""A linear regression model derived from a least-squares fit with an
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/stat/__init__.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/stat/__init__.py b/python/pyspark/mllib/stat/__init__.py
index b686d95..e3e1285 100644
--- a/python/pyspark/mllib/stat/__init__.py
+++ b/python/pyspark/mllib/stat/__init__.py
@@ -21,5 +21,7 @@ Python package for statistical functions in MLlib.
from pyspark.mllib.stat._statistics import *
from pyspark.mllib.stat.distribution import MultivariateGaussian
+from pyspark.mllib.stat.test import ChiSqTestResult
-__all__ = ["Statistics", "MultivariateStatisticalSummary", "MultivariateGaussian"]
+__all__ = ["Statistics", "MultivariateStatisticalSummary", "ChiSqTestResult",
+ "MultivariateGaussian"]
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/stat/distribution.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/stat/distribution.py b/python/pyspark/mllib/stat/distribution.py
index 07792e1..46f7a1d 100644
--- a/python/pyspark/mllib/stat/distribution.py
+++ b/python/pyspark/mllib/stat/distribution.py
@@ -22,7 +22,8 @@ __all__ = ['MultivariateGaussian']
class MultivariateGaussian(namedtuple('MultivariateGaussian', ['mu', 'sigma'])):
- """ Represents a (mu, sigma) tuple
+ """Represents a (mu, sigma) tuple
+
>>> m = MultivariateGaussian(Vectors.dense([11,12]),DenseMatrix(2, 2, (1.0, 3.0, 5.0, 2.0)))
>>> (m.mu, m.sigma.toArray())
(DenseVector([11.0, 12.0]), array([[ 1., 5.],[ 3., 2.]]))
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/python/pyspark/mllib/tree.py
----------------------------------------------------------------------
diff --git a/python/pyspark/mllib/tree.py b/python/pyspark/mllib/tree.py
index aae48f2..02d551b 100644
--- a/python/pyspark/mllib/tree.py
+++ b/python/pyspark/mllib/tree.py
@@ -20,12 +20,12 @@ from __future__ import absolute_import
import random
from pyspark import SparkContext, RDD
-from pyspark.mllib.common import callMLlibFunc, JavaModelWrapper
+from pyspark.mllib.common import callMLlibFunc, inherit_doc, JavaModelWrapper
from pyspark.mllib.linalg import _convert_to_vector
from pyspark.mllib.regression import LabeledPoint
__all__ = ['DecisionTreeModel', 'DecisionTree', 'RandomForestModel',
- 'RandomForest', 'GradientBoostedTrees']
+ 'RandomForest', 'GradientBoostedTreesModel', 'GradientBoostedTrees']
class TreeEnsembleModel(JavaModelWrapper):
@@ -216,6 +216,7 @@ class DecisionTree(object):
impurity, maxDepth, maxBins, minInstancesPerNode, minInfoGain)
+@inherit_doc
class RandomForestModel(TreeEnsembleModel):
"""
.. note:: Experimental
@@ -381,6 +382,7 @@ class RandomForest(object):
featureSubsetStrategy, impurity, maxDepth, maxBins, seed)
+@inherit_doc
class GradientBoostedTreesModel(TreeEnsembleModel):
"""
.. note:: Experimental
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@spark.apache.org
For additional commands, e-mail: commits-help@spark.apache.org
[2/2] spark git commit: [SPARK-5867] [SPARK-5892] [doc] [ml] [mllib]
Doc cleanups for 1.3 release
Posted by me...@apache.org.
[SPARK-5867] [SPARK-5892] [doc] [ml] [mllib] Doc cleanups for 1.3 release
For SPARK-5867:
* The spark.ml programming guide needs to be updated to use the new SQL DataFrame API instead of the old SchemaRDD API.
* It should also include Python examples now.
For SPARK-5892:
* Fix Python docs
* Various other cleanups
BTW, I accidentally merged this with master. If you want to compile it on your own, use this branch which is based on spark/branch-1.3 and cherry-picks the commits from this PR: [https://github.com/jkbradley/spark/tree/doc-review-1.3-check]
CC: mengxr (ML), davies (Python docs)
Author: Joseph K. Bradley <jo...@databricks.com>
Closes #4675 from jkbradley/doc-review-1.3 and squashes the following commits:
f191bb0 [Joseph K. Bradley] small cleanups
e786efa [Joseph K. Bradley] small doc corrections
6b1ab4a [Joseph K. Bradley] fixed python lint test
946affa [Joseph K. Bradley] Added sample data for ml.MovieLensALS example. Changed spark.ml Java examples to use DataFrames API instead of sql()
da81558 [Joseph K. Bradley] Merge remote-tracking branch 'upstream/master' into doc-review-1.3
629dbf5 [Joseph K. Bradley] Updated based on code review: * made new page for old migration guides * small fixes * moved inherit_doc in python
b9df7c4 [Joseph K. Bradley] Small cleanups: toDF to toDF(), adding s for string interpolation
34b067f [Joseph K. Bradley] small doc correction
da16aef [Joseph K. Bradley] Fixed python mllib docs
8cce91c [Joseph K. Bradley] GMM: removed old imports, added some doc
695f3f6 [Joseph K. Bradley] partly done trying to fix inherit_doc for class hierarchies in python docs
a72c018 [Joseph K. Bradley] made ChiSqTestResult appear in python docs
b05a80d [Joseph K. Bradley] organize imports. doc cleanups
e572827 [Joseph K. Bradley] updated programming guide for ml and mllib
Project: http://git-wip-us.apache.org/repos/asf/spark/repo
Commit: http://git-wip-us.apache.org/repos/asf/spark/commit/4a17eedb
Tree: http://git-wip-us.apache.org/repos/asf/spark/tree/4a17eedb
Diff: http://git-wip-us.apache.org/repos/asf/spark/diff/4a17eedb
Branch: refs/heads/master
Commit: 4a17eedb16343413e5b6f8bb58c6da8952ee7ab6
Parents: d3dfebe
Author: Joseph K. Bradley <jo...@databricks.com>
Authored: Fri Feb 20 02:31:32 2015 -0800
Committer: Xiangrui Meng <me...@databricks.com>
Committed: Fri Feb 20 02:31:32 2015 -0800
----------------------------------------------------------------------
data/mllib/als/sample_movielens_movies.txt | 100 ++
data/mllib/als/sample_movielens_ratings.txt | 1501 ++++++++++++++++++
docs/ml-guide.md | 249 +--
docs/mllib-guide.md | 133 +-
docs/mllib-migration-guides.md | 67 +
.../examples/ml/JavaCrossValidatorExample.java | 7 +-
.../examples/ml/JavaSimpleParamsExample.java | 8 +-
.../JavaSimpleTextClassificationPipeline.java | 7 +-
.../ml/simple_text_classification_pipeline.py | 4 +-
.../examples/ml/CrossValidatorExample.scala | 2 +-
.../apache/spark/examples/ml/MovieLensALS.scala | 32 +-
.../spark/examples/ml/SimpleParamsExample.scala | 6 +-
.../ml/SimpleTextClassificationPipeline.scala | 2 +-
.../mllib/clustering/GaussianMixture.scala | 12 +-
.../apache/spark/mllib/linalg/Matrices.scala | 6 +-
.../org/apache/spark/mllib/linalg/Vectors.scala | 2 +-
.../spark/mllib/optimization/Gradient.scala | 14 +-
python/docs/pyspark.mllib.rst | 11 +-
python/pyspark/ml/classification.py | 3 +-
python/pyspark/ml/feature.py | 3 +-
python/pyspark/ml/pipeline.py | 3 +-
python/pyspark/ml/util.py | 14 -
python/pyspark/ml/wrapper.py | 2 +-
python/pyspark/mllib/__init__.py | 2 +-
python/pyspark/mllib/clustering.py | 2 +-
python/pyspark/mllib/common.py | 17 +
python/pyspark/mllib/regression.py | 8 +-
python/pyspark/mllib/stat/__init__.py | 4 +-
python/pyspark/mllib/stat/distribution.py | 3 +-
python/pyspark/mllib/tree.py | 6 +-
30 files changed, 1940 insertions(+), 290 deletions(-)
----------------------------------------------------------------------
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/data/mllib/als/sample_movielens_movies.txt
----------------------------------------------------------------------
diff --git a/data/mllib/als/sample_movielens_movies.txt b/data/mllib/als/sample_movielens_movies.txt
new file mode 100644
index 0000000..934a025
--- /dev/null
+++ b/data/mllib/als/sample_movielens_movies.txt
@@ -0,0 +1,100 @@
+0::Movie 0::Romance|Comedy
+1::Movie 1::Action|Anime
+2::Movie 2::Romance|Thriller
+3::Movie 3::Action|Romance
+4::Movie 4::Anime|Comedy
+5::Movie 5::Action|Action
+6::Movie 6::Action|Comedy
+7::Movie 7::Anime|Comedy
+8::Movie 8::Comedy|Action
+9::Movie 9::Anime|Thriller
+10::Movie 10::Action|Anime
+11::Movie 11::Action|Anime
+12::Movie 12::Anime|Comedy
+13::Movie 13::Thriller|Action
+14::Movie 14::Anime|Comedy
+15::Movie 15::Comedy|Thriller
+16::Movie 16::Anime|Romance
+17::Movie 17::Thriller|Action
+18::Movie 18::Action|Comedy
+19::Movie 19::Anime|Romance
+20::Movie 20::Action|Anime
+21::Movie 21::Romance|Thriller
+22::Movie 22::Romance|Romance
+23::Movie 23::Comedy|Comedy
+24::Movie 24::Anime|Action
+25::Movie 25::Comedy|Comedy
+26::Movie 26::Anime|Romance
+27::Movie 27::Anime|Anime
+28::Movie 28::Thriller|Anime
+29::Movie 29::Anime|Romance
+30::Movie 30::Thriller|Romance
+31::Movie 31::Thriller|Romance
+32::Movie 32::Comedy|Anime
+33::Movie 33::Comedy|Comedy
+34::Movie 34::Anime|Anime
+35::Movie 35::Action|Thriller
+36::Movie 36::Anime|Romance
+37::Movie 37::Romance|Anime
+38::Movie 38::Thriller|Romance
+39::Movie 39::Romance|Comedy
+40::Movie 40::Action|Anime
+41::Movie 41::Comedy|Thriller
+42::Movie 42::Comedy|Action
+43::Movie 43::Thriller|Anime
+44::Movie 44::Anime|Action
+45::Movie 45::Comedy|Romance
+46::Movie 46::Comedy|Action
+47::Movie 47::Romance|Comedy
+48::Movie 48::Action|Comedy
+49::Movie 49::Romance|Romance
+50::Movie 50::Comedy|Romance
+51::Movie 51::Action|Action
+52::Movie 52::Thriller|Action
+53::Movie 53::Action|Action
+54::Movie 54::Romance|Thriller
+55::Movie 55::Anime|Romance
+56::Movie 56::Comedy|Action
+57::Movie 57::Action|Anime
+58::Movie 58::Thriller|Romance
+59::Movie 59::Thriller|Comedy
+60::Movie 60::Anime|Comedy
+61::Movie 61::Comedy|Action
+62::Movie 62::Comedy|Romance
+63::Movie 63::Romance|Thriller
+64::Movie 64::Romance|Action
+65::Movie 65::Anime|Romance
+66::Movie 66::Comedy|Action
+67::Movie 67::Thriller|Anime
+68::Movie 68::Thriller|Romance
+69::Movie 69::Action|Comedy
+70::Movie 70::Thriller|Thriller
+71::Movie 71::Action|Comedy
+72::Movie 72::Thriller|Romance
+73::Movie 73::Comedy|Action
+74::Movie 74::Action|Action
+75::Movie 75::Action|Action
+76::Movie 76::Comedy|Comedy
+77::Movie 77::Comedy|Comedy
+78::Movie 78::Comedy|Comedy
+79::Movie 79::Thriller|Thriller
+80::Movie 80::Comedy|Anime
+81::Movie 81::Comedy|Anime
+82::Movie 82::Romance|Anime
+83::Movie 83::Comedy|Thriller
+84::Movie 84::Anime|Action
+85::Movie 85::Thriller|Anime
+86::Movie 86::Romance|Anime
+87::Movie 87::Thriller|Thriller
+88::Movie 88::Romance|Thriller
+89::Movie 89::Action|Anime
+90::Movie 90::Anime|Romance
+91::Movie 91::Anime|Thriller
+92::Movie 92::Action|Comedy
+93::Movie 93::Romance|Thriller
+94::Movie 94::Thriller|Comedy
+95::Movie 95::Action|Action
+96::Movie 96::Thriller|Romance
+97::Movie 97::Thriller|Thriller
+98::Movie 98::Thriller|Comedy
+99::Movie 99::Thriller|Romance
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/data/mllib/als/sample_movielens_ratings.txt
----------------------------------------------------------------------
diff --git a/data/mllib/als/sample_movielens_ratings.txt b/data/mllib/als/sample_movielens_ratings.txt
new file mode 100644
index 0000000..0889142
--- /dev/null
+++ b/data/mllib/als/sample_movielens_ratings.txt
@@ -0,0 +1,1501 @@
+0::2::3::1424380312
+0::3::1::1424380312
+0::5::2::1424380312
+0::9::4::1424380312
+0::11::1::1424380312
+0::12::2::1424380312
+0::15::1::1424380312
+0::17::1::1424380312
+0::19::1::1424380312
+0::21::1::1424380312
+0::23::1::1424380312
+0::26::3::1424380312
+0::27::1::1424380312
+0::28::1::1424380312
+0::29::1::1424380312
+0::30::1::1424380312
+0::31::1::1424380312
+0::34::1::1424380312
+0::37::1::1424380312
+0::41::2::1424380312
+0::44::1::1424380312
+0::45::2::1424380312
+0::46::1::1424380312
+0::47::1::1424380312
+0::48::1::1424380312
+0::50::1::1424380312
+0::51::1::1424380312
+0::54::1::1424380312
+0::55::1::1424380312
+0::59::2::1424380312
+0::61::2::1424380312
+0::64::1::1424380312
+0::67::1::1424380312
+0::68::1::1424380312
+0::69::1::1424380312
+0::71::1::1424380312
+0::72::1::1424380312
+0::77::2::1424380312
+0::79::1::1424380312
+0::83::1::1424380312
+0::87::1::1424380312
+0::89::2::1424380312
+0::91::3::1424380312
+0::92::4::1424380312
+0::94::1::1424380312
+0::95::2::1424380312
+0::96::1::1424380312
+0::98::1::1424380312
+0::99::1::1424380312
+1::2::2::1424380312
+1::3::1::1424380312
+1::4::2::1424380312
+1::6::1::1424380312
+1::9::3::1424380312
+1::12::1::1424380312
+1::13::1::1424380312
+1::14::1::1424380312
+1::16::1::1424380312
+1::19::1::1424380312
+1::21::3::1424380312
+1::27::1::1424380312
+1::28::3::1424380312
+1::33::1::1424380312
+1::36::2::1424380312
+1::37::1::1424380312
+1::40::1::1424380312
+1::41::2::1424380312
+1::43::1::1424380312
+1::44::1::1424380312
+1::47::1::1424380312
+1::50::1::1424380312
+1::54::1::1424380312
+1::56::2::1424380312
+1::57::1::1424380312
+1::58::1::1424380312
+1::60::1::1424380312
+1::62::4::1424380312
+1::63::1::1424380312
+1::67::1::1424380312
+1::68::4::1424380312
+1::70::2::1424380312
+1::72::1::1424380312
+1::73::1::1424380312
+1::74::2::1424380312
+1::76::1::1424380312
+1::77::3::1424380312
+1::78::1::1424380312
+1::81::1::1424380312
+1::82::1::1424380312
+1::85::3::1424380312
+1::86::2::1424380312
+1::88::2::1424380312
+1::91::1::1424380312
+1::92::2::1424380312
+1::93::1::1424380312
+1::94::2::1424380312
+1::96::1::1424380312
+1::97::1::1424380312
+2::4::3::1424380312
+2::6::1::1424380312
+2::8::5::1424380312
+2::9::1::1424380312
+2::10::1::1424380312
+2::12::3::1424380312
+2::13::1::1424380312
+2::15::2::1424380312
+2::18::2::1424380312
+2::19::4::1424380312
+2::22::1::1424380312
+2::26::1::1424380312
+2::28::1::1424380312
+2::34::4::1424380312
+2::35::1::1424380312
+2::37::5::1424380312
+2::38::1::1424380312
+2::39::5::1424380312
+2::40::4::1424380312
+2::47::1::1424380312
+2::50::1::1424380312
+2::52::2::1424380312
+2::54::1::1424380312
+2::55::1::1424380312
+2::57::2::1424380312
+2::58::2::1424380312
+2::59::1::1424380312
+2::61::1::1424380312
+2::62::1::1424380312
+2::64::1::1424380312
+2::65::1::1424380312
+2::66::3::1424380312
+2::68::1::1424380312
+2::71::3::1424380312
+2::76::1::1424380312
+2::77::1::1424380312
+2::78::1::1424380312
+2::80::1::1424380312
+2::83::5::1424380312
+2::85::1::1424380312
+2::87::2::1424380312
+2::88::1::1424380312
+2::89::4::1424380312
+2::90::1::1424380312
+2::92::4::1424380312
+2::93::5::1424380312
+3::0::1::1424380312
+3::1::1::1424380312
+3::2::1::1424380312
+3::7::3::1424380312
+3::8::3::1424380312
+3::9::1::1424380312
+3::14::1::1424380312
+3::15::1::1424380312
+3::16::1::1424380312
+3::18::4::1424380312
+3::19::1::1424380312
+3::24::3::1424380312
+3::26::1::1424380312
+3::29::3::1424380312
+3::33::1::1424380312
+3::34::3::1424380312
+3::35::1::1424380312
+3::36::3::1424380312
+3::37::1::1424380312
+3::38::2::1424380312
+3::43::1::1424380312
+3::44::1::1424380312
+3::46::1::1424380312
+3::47::1::1424380312
+3::51::5::1424380312
+3::52::3::1424380312
+3::56::1::1424380312
+3::58::1::1424380312
+3::60::3::1424380312
+3::62::1::1424380312
+3::65::2::1424380312
+3::66::1::1424380312
+3::67::1::1424380312
+3::68::2::1424380312
+3::70::1::1424380312
+3::72::2::1424380312
+3::76::3::1424380312
+3::79::3::1424380312
+3::80::4::1424380312
+3::81::1::1424380312
+3::83::1::1424380312
+3::84::1::1424380312
+3::86::1::1424380312
+3::87::2::1424380312
+3::88::4::1424380312
+3::89::1::1424380312
+3::91::1::1424380312
+3::94::3::1424380312
+4::1::1::1424380312
+4::6::1::1424380312
+4::8::1::1424380312
+4::9::1::1424380312
+4::10::1::1424380312
+4::11::1::1424380312
+4::12::1::1424380312
+4::13::1::1424380312
+4::14::2::1424380312
+4::15::1::1424380312
+4::17::1::1424380312
+4::20::1::1424380312
+4::22::1::1424380312
+4::23::1::1424380312
+4::24::1::1424380312
+4::29::4::1424380312
+4::30::1::1424380312
+4::31::1::1424380312
+4::34::1::1424380312
+4::35::1::1424380312
+4::36::1::1424380312
+4::39::2::1424380312
+4::40::3::1424380312
+4::41::4::1424380312
+4::43::2::1424380312
+4::44::1::1424380312
+4::45::1::1424380312
+4::46::1::1424380312
+4::47::1::1424380312
+4::49::2::1424380312
+4::50::1::1424380312
+4::51::1::1424380312
+4::52::4::1424380312
+4::54::1::1424380312
+4::55::1::1424380312
+4::60::3::1424380312
+4::61::1::1424380312
+4::62::4::1424380312
+4::63::3::1424380312
+4::65::1::1424380312
+4::67::2::1424380312
+4::69::1::1424380312
+4::70::4::1424380312
+4::71::1::1424380312
+4::73::1::1424380312
+4::78::1::1424380312
+4::84::1::1424380312
+4::85::1::1424380312
+4::87::3::1424380312
+4::88::3::1424380312
+4::89::2::1424380312
+4::96::1::1424380312
+4::97::1::1424380312
+4::98::1::1424380312
+4::99::1::1424380312
+5::0::1::1424380312
+5::1::1::1424380312
+5::4::1::1424380312
+5::5::1::1424380312
+5::8::1::1424380312
+5::9::3::1424380312
+5::10::2::1424380312
+5::13::3::1424380312
+5::15::1::1424380312
+5::19::1::1424380312
+5::20::3::1424380312
+5::21::2::1424380312
+5::23::3::1424380312
+5::27::1::1424380312
+5::28::1::1424380312
+5::29::1::1424380312
+5::31::1::1424380312
+5::36::3::1424380312
+5::38::2::1424380312
+5::39::1::1424380312
+5::42::1::1424380312
+5::48::3::1424380312
+5::49::4::1424380312
+5::50::3::1424380312
+5::51::1::1424380312
+5::52::1::1424380312
+5::54::1::1424380312
+5::55::5::1424380312
+5::56::3::1424380312
+5::58::1::1424380312
+5::60::1::1424380312
+5::61::1::1424380312
+5::64::3::1424380312
+5::65::2::1424380312
+5::68::4::1424380312
+5::70::1::1424380312
+5::71::1::1424380312
+5::72::1::1424380312
+5::74::1::1424380312
+5::79::1::1424380312
+5::81::2::1424380312
+5::84::1::1424380312
+5::85::1::1424380312
+5::86::1::1424380312
+5::88::1::1424380312
+5::90::4::1424380312
+5::91::2::1424380312
+5::95::2::1424380312
+5::99::1::1424380312
+6::0::1::1424380312
+6::1::1::1424380312
+6::2::3::1424380312
+6::5::1::1424380312
+6::6::1::1424380312
+6::9::1::1424380312
+6::10::1::1424380312
+6::15::2::1424380312
+6::16::2::1424380312
+6::17::1::1424380312
+6::18::1::1424380312
+6::20::1::1424380312
+6::21::1::1424380312
+6::22::1::1424380312
+6::24::1::1424380312
+6::25::5::1424380312
+6::26::1::1424380312
+6::28::1::1424380312
+6::30::1::1424380312
+6::33::1::1424380312
+6::38::1::1424380312
+6::39::1::1424380312
+6::43::4::1424380312
+6::44::1::1424380312
+6::45::1::1424380312
+6::48::1::1424380312
+6::49::1::1424380312
+6::50::1::1424380312
+6::53::1::1424380312
+6::54::1::1424380312
+6::55::1::1424380312
+6::56::1::1424380312
+6::58::4::1424380312
+6::59::1::1424380312
+6::60::1::1424380312
+6::61::3::1424380312
+6::63::3::1424380312
+6::66::1::1424380312
+6::67::3::1424380312
+6::68::1::1424380312
+6::69::1::1424380312
+6::71::2::1424380312
+6::73::1::1424380312
+6::75::1::1424380312
+6::77::1::1424380312
+6::79::1::1424380312
+6::81::1::1424380312
+6::84::1::1424380312
+6::85::3::1424380312
+6::86::1::1424380312
+6::87::1::1424380312
+6::88::1::1424380312
+6::89::1::1424380312
+6::91::2::1424380312
+6::94::1::1424380312
+6::95::2::1424380312
+6::96::1::1424380312
+7::1::1::1424380312
+7::2::2::1424380312
+7::3::1::1424380312
+7::4::1::1424380312
+7::7::1::1424380312
+7::10::1::1424380312
+7::11::2::1424380312
+7::14::2::1424380312
+7::15::1::1424380312
+7::16::1::1424380312
+7::18::1::1424380312
+7::21::1::1424380312
+7::22::1::1424380312
+7::23::1::1424380312
+7::25::5::1424380312
+7::26::1::1424380312
+7::29::4::1424380312
+7::30::1::1424380312
+7::31::3::1424380312
+7::32::1::1424380312
+7::33::1::1424380312
+7::35::1::1424380312
+7::37::2::1424380312
+7::39::3::1424380312
+7::40::2::1424380312
+7::42::2::1424380312
+7::44::1::1424380312
+7::45::2::1424380312
+7::47::4::1424380312
+7::48::1::1424380312
+7::49::1::1424380312
+7::53::1::1424380312
+7::54::1::1424380312
+7::55::1::1424380312
+7::56::1::1424380312
+7::59::1::1424380312
+7::61::2::1424380312
+7::62::3::1424380312
+7::63::2::1424380312
+7::66::1::1424380312
+7::67::3::1424380312
+7::74::1::1424380312
+7::75::1::1424380312
+7::76::3::1424380312
+7::77::1::1424380312
+7::81::1::1424380312
+7::82::1::1424380312
+7::84::2::1424380312
+7::85::4::1424380312
+7::86::1::1424380312
+7::92::2::1424380312
+7::96::1::1424380312
+7::97::1::1424380312
+7::98::1::1424380312
+8::0::1::1424380312
+8::2::4::1424380312
+8::3::2::1424380312
+8::4::2::1424380312
+8::5::1::1424380312
+8::7::1::1424380312
+8::9::1::1424380312
+8::11::1::1424380312
+8::15::1::1424380312
+8::18::1::1424380312
+8::19::1::1424380312
+8::21::1::1424380312
+8::29::5::1424380312
+8::31::3::1424380312
+8::33::1::1424380312
+8::35::1::1424380312
+8::36::1::1424380312
+8::40::2::1424380312
+8::44::1::1424380312
+8::45::1::1424380312
+8::50::1::1424380312
+8::51::1::1424380312
+8::52::5::1424380312
+8::53::5::1424380312
+8::54::1::1424380312
+8::55::1::1424380312
+8::56::1::1424380312
+8::58::4::1424380312
+8::60::3::1424380312
+8::62::4::1424380312
+8::64::1::1424380312
+8::67::3::1424380312
+8::69::1::1424380312
+8::71::1::1424380312
+8::72::3::1424380312
+8::77::3::1424380312
+8::78::1::1424380312
+8::79::1::1424380312
+8::83::1::1424380312
+8::85::5::1424380312
+8::86::1::1424380312
+8::88::1::1424380312
+8::90::1::1424380312
+8::92::2::1424380312
+8::95::4::1424380312
+8::96::3::1424380312
+8::97::1::1424380312
+8::98::1::1424380312
+8::99::1::1424380312
+9::2::3::1424380312
+9::3::1::1424380312
+9::4::1::1424380312
+9::5::1::1424380312
+9::6::1::1424380312
+9::7::5::1424380312
+9::9::1::1424380312
+9::12::1::1424380312
+9::14::3::1424380312
+9::15::1::1424380312
+9::19::1::1424380312
+9::21::1::1424380312
+9::22::1::1424380312
+9::24::1::1424380312
+9::25::1::1424380312
+9::26::1::1424380312
+9::30::3::1424380312
+9::32::4::1424380312
+9::35::2::1424380312
+9::36::2::1424380312
+9::37::2::1424380312
+9::38::1::1424380312
+9::39::1::1424380312
+9::43::3::1424380312
+9::49::5::1424380312
+9::50::3::1424380312
+9::53::1::1424380312
+9::54::1::1424380312
+9::58::1::1424380312
+9::59::1::1424380312
+9::60::1::1424380312
+9::61::1::1424380312
+9::63::3::1424380312
+9::64::3::1424380312
+9::68::1::1424380312
+9::69::1::1424380312
+9::70::3::1424380312
+9::71::1::1424380312
+9::73::2::1424380312
+9::75::1::1424380312
+9::77::2::1424380312
+9::81::2::1424380312
+9::82::1::1424380312
+9::83::1::1424380312
+9::84::1::1424380312
+9::86::1::1424380312
+9::87::4::1424380312
+9::88::1::1424380312
+9::90::3::1424380312
+9::94::2::1424380312
+9::95::3::1424380312
+9::97::2::1424380312
+9::98::1::1424380312
+10::0::3::1424380312
+10::2::4::1424380312
+10::4::3::1424380312
+10::7::1::1424380312
+10::8::1::1424380312
+10::10::1::1424380312
+10::13::2::1424380312
+10::14::1::1424380312
+10::16::2::1424380312
+10::17::1::1424380312
+10::18::1::1424380312
+10::21::1::1424380312
+10::22::1::1424380312
+10::24::1::1424380312
+10::25::3::1424380312
+10::28::1::1424380312
+10::35::1::1424380312
+10::36::1::1424380312
+10::37::1::1424380312
+10::38::1::1424380312
+10::39::1::1424380312
+10::40::4::1424380312
+10::41::2::1424380312
+10::42::3::1424380312
+10::43::1::1424380312
+10::49::3::1424380312
+10::50::1::1424380312
+10::51::1::1424380312
+10::52::1::1424380312
+10::55::2::1424380312
+10::56::1::1424380312
+10::58::1::1424380312
+10::63::1::1424380312
+10::66::1::1424380312
+10::67::2::1424380312
+10::68::1::1424380312
+10::75::1::1424380312
+10::77::1::1424380312
+10::79::1::1424380312
+10::86::1::1424380312
+10::89::3::1424380312
+10::90::1::1424380312
+10::97::1::1424380312
+10::98::1::1424380312
+11::0::1::1424380312
+11::6::2::1424380312
+11::9::1::1424380312
+11::10::1::1424380312
+11::11::1::1424380312
+11::12::1::1424380312
+11::13::4::1424380312
+11::16::1::1424380312
+11::18::5::1424380312
+11::19::4::1424380312
+11::20::1::1424380312
+11::21::1::1424380312
+11::22::1::1424380312
+11::23::5::1424380312
+11::25::1::1424380312
+11::27::5::1424380312
+11::30::5::1424380312
+11::32::5::1424380312
+11::35::3::1424380312
+11::36::2::1424380312
+11::37::2::1424380312
+11::38::4::1424380312
+11::39::1::1424380312
+11::40::1::1424380312
+11::41::1::1424380312
+11::43::2::1424380312
+11::45::1::1424380312
+11::47::1::1424380312
+11::48::5::1424380312
+11::50::4::1424380312
+11::51::3::1424380312
+11::59::1::1424380312
+11::61::1::1424380312
+11::62::1::1424380312
+11::64::1::1424380312
+11::66::4::1424380312
+11::67::1::1424380312
+11::69::5::1424380312
+11::70::1::1424380312
+11::71::3::1424380312
+11::72::3::1424380312
+11::75::3::1424380312
+11::76::1::1424380312
+11::77::1::1424380312
+11::78::1::1424380312
+11::79::5::1424380312
+11::80::3::1424380312
+11::81::4::1424380312
+11::82::1::1424380312
+11::86::1::1424380312
+11::88::1::1424380312
+11::89::1::1424380312
+11::90::4::1424380312
+11::94::2::1424380312
+11::97::3::1424380312
+11::99::1::1424380312
+12::2::1::1424380312
+12::4::1::1424380312
+12::6::1::1424380312
+12::7::3::1424380312
+12::8::1::1424380312
+12::14::1::1424380312
+12::15::2::1424380312
+12::16::4::1424380312
+12::17::5::1424380312
+12::18::2::1424380312
+12::21::1::1424380312
+12::22::2::1424380312
+12::23::3::1424380312
+12::24::1::1424380312
+12::25::1::1424380312
+12::27::5::1424380312
+12::30::2::1424380312
+12::31::4::1424380312
+12::35::5::1424380312
+12::38::1::1424380312
+12::41::1::1424380312
+12::44::2::1424380312
+12::45::1::1424380312
+12::50::4::1424380312
+12::51::1::1424380312
+12::52::1::1424380312
+12::53::1::1424380312
+12::54::1::1424380312
+12::56::2::1424380312
+12::57::1::1424380312
+12::60::1::1424380312
+12::63::1::1424380312
+12::64::5::1424380312
+12::66::3::1424380312
+12::67::1::1424380312
+12::70::1::1424380312
+12::72::1::1424380312
+12::74::1::1424380312
+12::75::1::1424380312
+12::77::1::1424380312
+12::78::1::1424380312
+12::79::3::1424380312
+12::82::2::1424380312
+12::83::1::1424380312
+12::84::1::1424380312
+12::85::1::1424380312
+12::86::1::1424380312
+12::87::1::1424380312
+12::88::1::1424380312
+12::91::3::1424380312
+12::92::1::1424380312
+12::94::4::1424380312
+12::95::2::1424380312
+12::96::1::1424380312
+12::98::2::1424380312
+13::0::1::1424380312
+13::3::1::1424380312
+13::4::2::1424380312
+13::5::1::1424380312
+13::6::1::1424380312
+13::12::1::1424380312
+13::14::2::1424380312
+13::15::1::1424380312
+13::17::1::1424380312
+13::18::3::1424380312
+13::20::1::1424380312
+13::21::1::1424380312
+13::22::1::1424380312
+13::26::1::1424380312
+13::27::1::1424380312
+13::29::3::1424380312
+13::31::1::1424380312
+13::33::1::1424380312
+13::40::2::1424380312
+13::43::2::1424380312
+13::44::1::1424380312
+13::45::1::1424380312
+13::49::1::1424380312
+13::51::1::1424380312
+13::52::2::1424380312
+13::53::3::1424380312
+13::54::1::1424380312
+13::62::1::1424380312
+13::63::2::1424380312
+13::64::1::1424380312
+13::68::1::1424380312
+13::71::1::1424380312
+13::72::3::1424380312
+13::73::1::1424380312
+13::74::3::1424380312
+13::77::2::1424380312
+13::78::1::1424380312
+13::79::2::1424380312
+13::83::3::1424380312
+13::85::1::1424380312
+13::86::1::1424380312
+13::87::2::1424380312
+13::88::2::1424380312
+13::90::1::1424380312
+13::93::4::1424380312
+13::94::1::1424380312
+13::98::1::1424380312
+13::99::1::1424380312
+14::1::1::1424380312
+14::3::3::1424380312
+14::4::1::1424380312
+14::5::1::1424380312
+14::6::1::1424380312
+14::7::1::1424380312
+14::9::1::1424380312
+14::10::1::1424380312
+14::11::1::1424380312
+14::12::1::1424380312
+14::13::1::1424380312
+14::14::3::1424380312
+14::15::1::1424380312
+14::16::1::1424380312
+14::17::1::1424380312
+14::20::1::1424380312
+14::21::1::1424380312
+14::24::1::1424380312
+14::25::2::1424380312
+14::27::1::1424380312
+14::28::1::1424380312
+14::29::5::1424380312
+14::31::3::1424380312
+14::34::1::1424380312
+14::36::1::1424380312
+14::37::2::1424380312
+14::39::2::1424380312
+14::40::1::1424380312
+14::44::1::1424380312
+14::45::1::1424380312
+14::47::3::1424380312
+14::48::1::1424380312
+14::49::1::1424380312
+14::51::1::1424380312
+14::52::5::1424380312
+14::53::3::1424380312
+14::54::1::1424380312
+14::55::1::1424380312
+14::56::1::1424380312
+14::62::4::1424380312
+14::63::5::1424380312
+14::67::3::1424380312
+14::68::1::1424380312
+14::69::3::1424380312
+14::71::1::1424380312
+14::72::4::1424380312
+14::73::1::1424380312
+14::76::5::1424380312
+14::79::1::1424380312
+14::82::1::1424380312
+14::83::1::1424380312
+14::88::1::1424380312
+14::93::3::1424380312
+14::94::1::1424380312
+14::95::2::1424380312
+14::96::4::1424380312
+14::98::1::1424380312
+15::0::1::1424380312
+15::1::4::1424380312
+15::2::1::1424380312
+15::5::2::1424380312
+15::6::1::1424380312
+15::7::1::1424380312
+15::13::1::1424380312
+15::14::1::1424380312
+15::15::1::1424380312
+15::17::2::1424380312
+15::19::2::1424380312
+15::22::2::1424380312
+15::23::2::1424380312
+15::25::1::1424380312
+15::26::3::1424380312
+15::27::1::1424380312
+15::28::2::1424380312
+15::29::1::1424380312
+15::32::1::1424380312
+15::33::2::1424380312
+15::34::1::1424380312
+15::35::2::1424380312
+15::36::1::1424380312
+15::37::1::1424380312
+15::39::1::1424380312
+15::42::1::1424380312
+15::46::5::1424380312
+15::48::2::1424380312
+15::50::2::1424380312
+15::51::1::1424380312
+15::52::1::1424380312
+15::58::1::1424380312
+15::62::1::1424380312
+15::64::3::1424380312
+15::65::2::1424380312
+15::72::1::1424380312
+15::73::1::1424380312
+15::74::1::1424380312
+15::79::1::1424380312
+15::80::1::1424380312
+15::81::1::1424380312
+15::82::2::1424380312
+15::85::1::1424380312
+15::87::1::1424380312
+15::91::2::1424380312
+15::96::1::1424380312
+15::97::1::1424380312
+15::98::3::1424380312
+16::2::1::1424380312
+16::5::3::1424380312
+16::6::2::1424380312
+16::7::1::1424380312
+16::9::1::1424380312
+16::12::1::1424380312
+16::14::1::1424380312
+16::15::1::1424380312
+16::19::1::1424380312
+16::21::2::1424380312
+16::29::4::1424380312
+16::30::2::1424380312
+16::32::1::1424380312
+16::34::1::1424380312
+16::36::1::1424380312
+16::38::1::1424380312
+16::46::1::1424380312
+16::47::3::1424380312
+16::48::1::1424380312
+16::49::1::1424380312
+16::50::1::1424380312
+16::51::5::1424380312
+16::54::5::1424380312
+16::55::1::1424380312
+16::56::2::1424380312
+16::57::1::1424380312
+16::60::1::1424380312
+16::63::2::1424380312
+16::65::1::1424380312
+16::67::1::1424380312
+16::72::1::1424380312
+16::74::1::1424380312
+16::80::1::1424380312
+16::81::1::1424380312
+16::82::1::1424380312
+16::85::5::1424380312
+16::86::1::1424380312
+16::90::5::1424380312
+16::91::1::1424380312
+16::93::1::1424380312
+16::94::3::1424380312
+16::95::2::1424380312
+16::96::3::1424380312
+16::98::3::1424380312
+16::99::1::1424380312
+17::2::1::1424380312
+17::3::1::1424380312
+17::6::1::1424380312
+17::10::4::1424380312
+17::11::1::1424380312
+17::13::2::1424380312
+17::17::5::1424380312
+17::19::1::1424380312
+17::20::5::1424380312
+17::22::4::1424380312
+17::28::1::1424380312
+17::29::1::1424380312
+17::33::1::1424380312
+17::34::1::1424380312
+17::35::2::1424380312
+17::37::1::1424380312
+17::38::1::1424380312
+17::45::1::1424380312
+17::46::5::1424380312
+17::47::1::1424380312
+17::49::3::1424380312
+17::51::1::1424380312
+17::55::5::1424380312
+17::56::3::1424380312
+17::57::1::1424380312
+17::58::1::1424380312
+17::59::1::1424380312
+17::60::1::1424380312
+17::63::1::1424380312
+17::66::1::1424380312
+17::68::4::1424380312
+17::69::1::1424380312
+17::70::1::1424380312
+17::72::1::1424380312
+17::73::3::1424380312
+17::78::1::1424380312
+17::79::1::1424380312
+17::82::2::1424380312
+17::84::1::1424380312
+17::90::5::1424380312
+17::91::3::1424380312
+17::92::1::1424380312
+17::93::1::1424380312
+17::94::4::1424380312
+17::95::2::1424380312
+17::97::1::1424380312
+18::1::1::1424380312
+18::4::3::1424380312
+18::5::2::1424380312
+18::6::1::1424380312
+18::7::1::1424380312
+18::10::1::1424380312
+18::11::4::1424380312
+18::12::2::1424380312
+18::13::1::1424380312
+18::15::1::1424380312
+18::18::1::1424380312
+18::20::1::1424380312
+18::21::2::1424380312
+18::22::1::1424380312
+18::23::2::1424380312
+18::25::1::1424380312
+18::26::1::1424380312
+18::27::1::1424380312
+18::28::5::1424380312
+18::29::1::1424380312
+18::31::1::1424380312
+18::32::1::1424380312
+18::36::1::1424380312
+18::38::5::1424380312
+18::39::5::1424380312
+18::40::1::1424380312
+18::42::1::1424380312
+18::43::1::1424380312
+18::44::4::1424380312
+18::46::1::1424380312
+18::47::1::1424380312
+18::48::1::1424380312
+18::51::2::1424380312
+18::55::1::1424380312
+18::56::1::1424380312
+18::57::1::1424380312
+18::62::1::1424380312
+18::63::1::1424380312
+18::66::3::1424380312
+18::67::1::1424380312
+18::70::1::1424380312
+18::75::1::1424380312
+18::76::3::1424380312
+18::77::1::1424380312
+18::80::3::1424380312
+18::81::3::1424380312
+18::82::1::1424380312
+18::83::5::1424380312
+18::84::1::1424380312
+18::97::1::1424380312
+18::98::1::1424380312
+18::99::2::1424380312
+19::0::1::1424380312
+19::1::1::1424380312
+19::2::1::1424380312
+19::4::1::1424380312
+19::6::2::1424380312
+19::11::1::1424380312
+19::12::1::1424380312
+19::14::1::1424380312
+19::23::1::1424380312
+19::26::1::1424380312
+19::31::1::1424380312
+19::32::4::1424380312
+19::33::1::1424380312
+19::34::1::1424380312
+19::37::1::1424380312
+19::38::1::1424380312
+19::41::1::1424380312
+19::43::1::1424380312
+19::45::1::1424380312
+19::48::1::1424380312
+19::49::1::1424380312
+19::50::2::1424380312
+19::53::2::1424380312
+19::54::3::1424380312
+19::55::1::1424380312
+19::56::2::1424380312
+19::58::1::1424380312
+19::61::1::1424380312
+19::62::1::1424380312
+19::63::1::1424380312
+19::64::1::1424380312
+19::65::1::1424380312
+19::69::2::1424380312
+19::72::1::1424380312
+19::74::3::1424380312
+19::76::1::1424380312
+19::78::1::1424380312
+19::79::1::1424380312
+19::81::1::1424380312
+19::82::1::1424380312
+19::84::1::1424380312
+19::86::1::1424380312
+19::87::2::1424380312
+19::90::4::1424380312
+19::93::1::1424380312
+19::94::4::1424380312
+19::95::2::1424380312
+19::96::1::1424380312
+19::98::4::1424380312
+20::0::1::1424380312
+20::1::1::1424380312
+20::2::2::1424380312
+20::4::2::1424380312
+20::6::1::1424380312
+20::8::1::1424380312
+20::12::1::1424380312
+20::21::2::1424380312
+20::22::5::1424380312
+20::24::2::1424380312
+20::25::1::1424380312
+20::26::1::1424380312
+20::29::2::1424380312
+20::30::2::1424380312
+20::32::2::1424380312
+20::39::1::1424380312
+20::40::1::1424380312
+20::41::2::1424380312
+20::45::2::1424380312
+20::48::1::1424380312
+20::50::1::1424380312
+20::51::3::1424380312
+20::53::3::1424380312
+20::55::1::1424380312
+20::57::2::1424380312
+20::60::1::1424380312
+20::61::1::1424380312
+20::64::1::1424380312
+20::66::1::1424380312
+20::70::2::1424380312
+20::72::1::1424380312
+20::73::2::1424380312
+20::75::4::1424380312
+20::76::1::1424380312
+20::77::4::1424380312
+20::78::1::1424380312
+20::79::1::1424380312
+20::84::2::1424380312
+20::85::2::1424380312
+20::88::3::1424380312
+20::89::1::1424380312
+20::90::3::1424380312
+20::91::1::1424380312
+20::92::2::1424380312
+20::93::1::1424380312
+20::94::4::1424380312
+20::97::1::1424380312
+21::0::1::1424380312
+21::2::4::1424380312
+21::3::1::1424380312
+21::7::2::1424380312
+21::11::1::1424380312
+21::12::1::1424380312
+21::13::1::1424380312
+21::14::3::1424380312
+21::17::1::1424380312
+21::19::1::1424380312
+21::20::1::1424380312
+21::21::1::1424380312
+21::22::1::1424380312
+21::23::1::1424380312
+21::24::1::1424380312
+21::27::1::1424380312
+21::29::5::1424380312
+21::30::2::1424380312
+21::38::1::1424380312
+21::40::2::1424380312
+21::43::3::1424380312
+21::44::1::1424380312
+21::45::1::1424380312
+21::46::1::1424380312
+21::48::1::1424380312
+21::51::1::1424380312
+21::53::5::1424380312
+21::54::1::1424380312
+21::55::1::1424380312
+21::56::1::1424380312
+21::58::3::1424380312
+21::59::3::1424380312
+21::64::1::1424380312
+21::66::1::1424380312
+21::68::1::1424380312
+21::71::1::1424380312
+21::73::1::1424380312
+21::74::4::1424380312
+21::80::1::1424380312
+21::81::1::1424380312
+21::83::1::1424380312
+21::84::1::1424380312
+21::85::3::1424380312
+21::87::4::1424380312
+21::89::2::1424380312
+21::92::2::1424380312
+21::96::3::1424380312
+21::99::1::1424380312
+22::0::1::1424380312
+22::3::2::1424380312
+22::5::2::1424380312
+22::6::2::1424380312
+22::9::1::1424380312
+22::10::1::1424380312
+22::11::1::1424380312
+22::13::1::1424380312
+22::14::1::1424380312
+22::16::1::1424380312
+22::18::3::1424380312
+22::19::1::1424380312
+22::22::5::1424380312
+22::25::1::1424380312
+22::26::1::1424380312
+22::29::3::1424380312
+22::30::5::1424380312
+22::32::4::1424380312
+22::33::1::1424380312
+22::35::1::1424380312
+22::36::3::1424380312
+22::37::1::1424380312
+22::40::1::1424380312
+22::41::3::1424380312
+22::44::1::1424380312
+22::45::2::1424380312
+22::48::1::1424380312
+22::51::5::1424380312
+22::55::1::1424380312
+22::56::2::1424380312
+22::60::3::1424380312
+22::61::1::1424380312
+22::62::4::1424380312
+22::63::1::1424380312
+22::65::1::1424380312
+22::66::1::1424380312
+22::68::4::1424380312
+22::69::4::1424380312
+22::70::3::1424380312
+22::71::1::1424380312
+22::74::5::1424380312
+22::75::5::1424380312
+22::78::1::1424380312
+22::80::3::1424380312
+22::81::1::1424380312
+22::82::1::1424380312
+22::84::1::1424380312
+22::86::1::1424380312
+22::87::3::1424380312
+22::88::5::1424380312
+22::90::2::1424380312
+22::92::3::1424380312
+22::95::2::1424380312
+22::96::2::1424380312
+22::98::4::1424380312
+22::99::1::1424380312
+23::0::1::1424380312
+23::2::1::1424380312
+23::4::1::1424380312
+23::6::2::1424380312
+23::10::4::1424380312
+23::12::1::1424380312
+23::13::4::1424380312
+23::14::1::1424380312
+23::15::1::1424380312
+23::18::4::1424380312
+23::22::2::1424380312
+23::23::4::1424380312
+23::24::1::1424380312
+23::25::1::1424380312
+23::26::1::1424380312
+23::27::5::1424380312
+23::28::1::1424380312
+23::29::1::1424380312
+23::30::4::1424380312
+23::32::5::1424380312
+23::33::2::1424380312
+23::36::3::1424380312
+23::37::1::1424380312
+23::38::1::1424380312
+23::39::1::1424380312
+23::43::1::1424380312
+23::48::5::1424380312
+23::49::5::1424380312
+23::50::4::1424380312
+23::53::1::1424380312
+23::55::5::1424380312
+23::57::1::1424380312
+23::59::1::1424380312
+23::60::1::1424380312
+23::61::1::1424380312
+23::64::4::1424380312
+23::65::5::1424380312
+23::66::2::1424380312
+23::67::1::1424380312
+23::68::3::1424380312
+23::69::1::1424380312
+23::72::1::1424380312
+23::73::3::1424380312
+23::77::1::1424380312
+23::82::2::1424380312
+23::83::1::1424380312
+23::84::1::1424380312
+23::85::1::1424380312
+23::87::3::1424380312
+23::88::1::1424380312
+23::95::2::1424380312
+23::97::1::1424380312
+24::4::1::1424380312
+24::6::3::1424380312
+24::7::1::1424380312
+24::10::2::1424380312
+24::12::1::1424380312
+24::15::1::1424380312
+24::19::1::1424380312
+24::24::1::1424380312
+24::27::3::1424380312
+24::30::5::1424380312
+24::31::1::1424380312
+24::32::3::1424380312
+24::33::1::1424380312
+24::37::1::1424380312
+24::39::1::1424380312
+24::40::1::1424380312
+24::42::1::1424380312
+24::43::3::1424380312
+24::45::2::1424380312
+24::46::1::1424380312
+24::47::1::1424380312
+24::48::1::1424380312
+24::49::1::1424380312
+24::50::1::1424380312
+24::52::5::1424380312
+24::57::1::1424380312
+24::59::4::1424380312
+24::63::4::1424380312
+24::65::1::1424380312
+24::66::1::1424380312
+24::67::1::1424380312
+24::68::3::1424380312
+24::69::5::1424380312
+24::71::1::1424380312
+24::72::4::1424380312
+24::77::4::1424380312
+24::78::1::1424380312
+24::80::1::1424380312
+24::82::1::1424380312
+24::84::1::1424380312
+24::86::1::1424380312
+24::87::1::1424380312
+24::88::2::1424380312
+24::89::1::1424380312
+24::90::5::1424380312
+24::91::1::1424380312
+24::92::1::1424380312
+24::94::2::1424380312
+24::95::1::1424380312
+24::96::5::1424380312
+24::98::1::1424380312
+24::99::1::1424380312
+25::1::3::1424380312
+25::2::1::1424380312
+25::7::1::1424380312
+25::9::1::1424380312
+25::12::3::1424380312
+25::16::3::1424380312
+25::17::1::1424380312
+25::18::1::1424380312
+25::20::1::1424380312
+25::22::1::1424380312
+25::23::1::1424380312
+25::26::2::1424380312
+25::29::1::1424380312
+25::30::1::1424380312
+25::31::2::1424380312
+25::33::4::1424380312
+25::34::3::1424380312
+25::35::2::1424380312
+25::36::1::1424380312
+25::37::1::1424380312
+25::40::1::1424380312
+25::41::1::1424380312
+25::43::1::1424380312
+25::47::4::1424380312
+25::50::1::1424380312
+25::51::1::1424380312
+25::53::1::1424380312
+25::56::1::1424380312
+25::58::2::1424380312
+25::64::2::1424380312
+25::67::2::1424380312
+25::68::1::1424380312
+25::70::1::1424380312
+25::71::4::1424380312
+25::73::1::1424380312
+25::74::1::1424380312
+25::76::1::1424380312
+25::79::1::1424380312
+25::82::1::1424380312
+25::84::2::1424380312
+25::85::1::1424380312
+25::91::3::1424380312
+25::92::1::1424380312
+25::94::1::1424380312
+25::95::1::1424380312
+25::97::2::1424380312
+26::0::1::1424380312
+26::1::1::1424380312
+26::2::1::1424380312
+26::3::1::1424380312
+26::4::4::1424380312
+26::5::2::1424380312
+26::6::3::1424380312
+26::7::5::1424380312
+26::13::3::1424380312
+26::14::1::1424380312
+26::16::1::1424380312
+26::18::3::1424380312
+26::20::1::1424380312
+26::21::3::1424380312
+26::22::5::1424380312
+26::23::5::1424380312
+26::24::5::1424380312
+26::27::1::1424380312
+26::31::1::1424380312
+26::35::1::1424380312
+26::36::4::1424380312
+26::40::1::1424380312
+26::44::1::1424380312
+26::45::2::1424380312
+26::47::1::1424380312
+26::48::1::1424380312
+26::49::3::1424380312
+26::50::2::1424380312
+26::52::1::1424380312
+26::54::4::1424380312
+26::55::1::1424380312
+26::57::3::1424380312
+26::58::1::1424380312
+26::61::1::1424380312
+26::62::2::1424380312
+26::66::1::1424380312
+26::68::4::1424380312
+26::71::1::1424380312
+26::73::4::1424380312
+26::76::1::1424380312
+26::81::3::1424380312
+26::85::1::1424380312
+26::86::3::1424380312
+26::88::5::1424380312
+26::91::1::1424380312
+26::94::5::1424380312
+26::95::1::1424380312
+26::96::1::1424380312
+26::97::1::1424380312
+27::0::1::1424380312
+27::9::1::1424380312
+27::10::1::1424380312
+27::18::4::1424380312
+27::19::3::1424380312
+27::20::1::1424380312
+27::22::2::1424380312
+27::24::2::1424380312
+27::25::1::1424380312
+27::27::3::1424380312
+27::28::1::1424380312
+27::29::1::1424380312
+27::31::1::1424380312
+27::33::3::1424380312
+27::40::1::1424380312
+27::42::1::1424380312
+27::43::1::1424380312
+27::44::3::1424380312
+27::45::1::1424380312
+27::51::3::1424380312
+27::52::1::1424380312
+27::55::3::1424380312
+27::57::1::1424380312
+27::59::1::1424380312
+27::60::1::1424380312
+27::61::1::1424380312
+27::64::1::1424380312
+27::66::3::1424380312
+27::68::1::1424380312
+27::70::1::1424380312
+27::71::2::1424380312
+27::72::1::1424380312
+27::75::3::1424380312
+27::78::1::1424380312
+27::80::3::1424380312
+27::82::1::1424380312
+27::83::3::1424380312
+27::86::1::1424380312
+27::87::2::1424380312
+27::90::1::1424380312
+27::91::1::1424380312
+27::92::1::1424380312
+27::93::1::1424380312
+27::94::2::1424380312
+27::95::1::1424380312
+27::98::1::1424380312
+28::0::3::1424380312
+28::1::1::1424380312
+28::2::4::1424380312
+28::3::1::1424380312
+28::6::1::1424380312
+28::7::1::1424380312
+28::12::5::1424380312
+28::13::2::1424380312
+28::14::1::1424380312
+28::15::1::1424380312
+28::17::1::1424380312
+28::19::3::1424380312
+28::20::1::1424380312
+28::23::3::1424380312
+28::24::3::1424380312
+28::27::1::1424380312
+28::29::1::1424380312
+28::33::1::1424380312
+28::34::1::1424380312
+28::36::1::1424380312
+28::38::2::1424380312
+28::39::2::1424380312
+28::44::1::1424380312
+28::45::1::1424380312
+28::49::4::1424380312
+28::50::1::1424380312
+28::52::1::1424380312
+28::54::1::1424380312
+28::56::1::1424380312
+28::57::3::1424380312
+28::58::1::1424380312
+28::59::1::1424380312
+28::60::1::1424380312
+28::62::3::1424380312
+28::63::1::1424380312
+28::65::1::1424380312
+28::75::1::1424380312
+28::78::1::1424380312
+28::81::5::1424380312
+28::82::4::1424380312
+28::83::1::1424380312
+28::85::1::1424380312
+28::88::2::1424380312
+28::89::4::1424380312
+28::90::1::1424380312
+28::92::5::1424380312
+28::94::1::1424380312
+28::95::2::1424380312
+28::98::1::1424380312
+28::99::1::1424380312
+29::3::1::1424380312
+29::4::1::1424380312
+29::5::1::1424380312
+29::7::2::1424380312
+29::9::1::1424380312
+29::10::3::1424380312
+29::11::1::1424380312
+29::13::3::1424380312
+29::14::1::1424380312
+29::15::1::1424380312
+29::17::3::1424380312
+29::19::3::1424380312
+29::22::3::1424380312
+29::23::4::1424380312
+29::25::1::1424380312
+29::29::1::1424380312
+29::31::1::1424380312
+29::32::4::1424380312
+29::33::2::1424380312
+29::36::2::1424380312
+29::38::3::1424380312
+29::39::1::1424380312
+29::42::1::1424380312
+29::46::5::1424380312
+29::49::3::1424380312
+29::51::2::1424380312
+29::59::1::1424380312
+29::61::1::1424380312
+29::62::1::1424380312
+29::67::1::1424380312
+29::68::3::1424380312
+29::69::1::1424380312
+29::70::1::1424380312
+29::74::1::1424380312
+29::75::1::1424380312
+29::79::2::1424380312
+29::80::1::1424380312
+29::81::2::1424380312
+29::83::1::1424380312
+29::85::1::1424380312
+29::86::1::1424380312
+29::90::4::1424380312
+29::93::1::1424380312
+29::94::4::1424380312
+29::97::1::1424380312
+29::99::1::1424380312
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/docs/ml-guide.md
----------------------------------------------------------------------
diff --git a/docs/ml-guide.md b/docs/ml-guide.md
index 4bf14fb..da6aef7 100644
--- a/docs/ml-guide.md
+++ b/docs/ml-guide.md
@@ -23,13 +23,13 @@ to `spark.ml`.
Spark ML standardizes APIs for machine learning algorithms to make it easier to combine multiple algorithms into a single pipeline, or workflow. This section covers the key concepts introduced by the Spark ML API.
-* **[ML Dataset](ml-guide.html#ml-dataset)**: Spark ML uses the [`SchemaRDD`](api/scala/index.html#org.apache.spark.sql.SchemaRDD) from Spark SQL as a dataset which can hold a variety of data types.
+* **[ML Dataset](ml-guide.html#ml-dataset)**: Spark ML uses the [`DataFrame`](api/scala/index.html#org.apache.spark.sql.DataFrame) from Spark SQL as a dataset which can hold a variety of data types.
E.g., a dataset could have different columns storing text, feature vectors, true labels, and predictions.
-* **[`Transformer`](ml-guide.html#transformers)**: A `Transformer` is an algorithm which can transform one `SchemaRDD` into another `SchemaRDD`.
+* **[`Transformer`](ml-guide.html#transformers)**: A `Transformer` is an algorithm which can transform one `DataFrame` into another `DataFrame`.
E.g., an ML model is a `Transformer` which transforms an RDD with features into an RDD with predictions.
-* **[`Estimator`](ml-guide.html#estimators)**: An `Estimator` is an algorithm which can be fit on a `SchemaRDD` to produce a `Transformer`.
+* **[`Estimator`](ml-guide.html#estimators)**: An `Estimator` is an algorithm which can be fit on a `DataFrame` to produce a `Transformer`.
E.g., a learning algorithm is an `Estimator` which trains on a dataset and produces a model.
* **[`Pipeline`](ml-guide.html#pipeline)**: A `Pipeline` chains multiple `Transformer`s and `Estimator`s together to specify an ML workflow.
@@ -39,20 +39,20 @@ E.g., a learning algorithm is an `Estimator` which trains on a dataset and produ
## ML Dataset
Machine learning can be applied to a wide variety of data types, such as vectors, text, images, and structured data.
-Spark ML adopts the [`SchemaRDD`](api/scala/index.html#org.apache.spark.sql.SchemaRDD) from Spark SQL in order to support a variety of data types under a unified Dataset concept.
+Spark ML adopts the [`DataFrame`](api/scala/index.html#org.apache.spark.sql.DataFrame) from Spark SQL in order to support a variety of data types under a unified Dataset concept.
-`SchemaRDD` supports many basic and structured types; see the [Spark SQL datatype reference](sql-programming-guide.html#spark-sql-datatype-reference) for a list of supported types.
-In addition to the types listed in the Spark SQL guide, `SchemaRDD` can use ML [`Vector`](api/scala/index.html#org.apache.spark.mllib.linalg.Vector) types.
+`DataFrame` supports many basic and structured types; see the [Spark SQL datatype reference](sql-programming-guide.html#spark-sql-datatype-reference) for a list of supported types.
+In addition to the types listed in the Spark SQL guide, `DataFrame` can use ML [`Vector`](api/scala/index.html#org.apache.spark.mllib.linalg.Vector) types.
-A `SchemaRDD` can be created either implicitly or explicitly from a regular `RDD`. See the code examples below and the [Spark SQL programming guide](sql-programming-guide.html) for examples.
+A `DataFrame` can be created either implicitly or explicitly from a regular `RDD`. See the code examples below and the [Spark SQL programming guide](sql-programming-guide.html) for examples.
-Columns in a `SchemaRDD` are named. The code examples below use names such as "text," "features," and "label."
+Columns in a `DataFrame` are named. The code examples below use names such as "text," "features," and "label."
## ML Algorithms
### Transformers
-A [`Transformer`](api/scala/index.html#org.apache.spark.ml.Transformer) is an abstraction which includes feature transformers and learned models. Technically, a `Transformer` implements a method `transform()` which converts one `SchemaRDD` into another, generally by appending one or more columns.
+A [`Transformer`](api/scala/index.html#org.apache.spark.ml.Transformer) is an abstraction which includes feature transformers and learned models. Technically, a `Transformer` implements a method `transform()` which converts one `DataFrame` into another, generally by appending one or more columns.
For example:
* A feature transformer might take a dataset, read a column (e.g., text), convert it into a new column (e.g., feature vectors), append the new column to the dataset, and output the updated dataset.
@@ -60,7 +60,7 @@ For example:
### Estimators
-An [`Estimator`](api/scala/index.html#org.apache.spark.ml.Estimator) abstracts the concept of a learning algorithm or any algorithm which fits or trains on data. Technically, an `Estimator` implements a method `fit()` which accepts a `SchemaRDD` and produces a `Transformer`.
+An [`Estimator`](api/scala/index.html#org.apache.spark.ml.Estimator) abstracts the concept of a learning algorithm or any algorithm which fits or trains on data. Technically, an `Estimator` implements a method `fit()` which accepts a `DataFrame` and produces a `Transformer`.
For example, a learning algorithm such as `LogisticRegression` is an `Estimator`, and calling `fit()` trains a `LogisticRegressionModel`, which is a `Transformer`.
### Properties of ML Algorithms
@@ -101,7 +101,7 @@ We illustrate this for the simple text document workflow. The figure below is f
Above, the top row represents a `Pipeline` with three stages.
The first two (`Tokenizer` and `HashingTF`) are `Transformer`s (blue), and the third (`LogisticRegression`) is an `Estimator` (red).
-The bottom row represents data flowing through the pipeline, where cylinders indicate `SchemaRDD`s.
+The bottom row represents data flowing through the pipeline, where cylinders indicate `DataFrame`s.
The `Pipeline.fit()` method is called on the original dataset which has raw text documents and labels.
The `Tokenizer.transform()` method splits the raw text documents into words, adding a new column with words into the dataset.
The `HashingTF.transform()` method converts the words column into feature vectors, adding a new column with those vectors to the dataset.
@@ -130,7 +130,7 @@ Each stage's `transform()` method updates the dataset and passes it to the next
*DAG `Pipeline`s*: A `Pipeline`'s stages are specified as an ordered array. The examples given here are all for linear `Pipeline`s, i.e., `Pipeline`s in which each stage uses data produced by the previous stage. It is possible to create non-linear `Pipeline`s as long as the data flow graph forms a Directed Acyclic Graph (DAG). This graph is currently specified implicitly based on the input and output column names of each stage (generally specified as parameters). If the `Pipeline` forms a DAG, then the stages must be specified in topological order.
-*Runtime checking*: Since `Pipeline`s can operate on datasets with varied types, they cannot use compile-time type checking. `Pipeline`s and `PipelineModel`s instead do runtime checking before actually running the `Pipeline`. This type checking is done using the dataset *schema*, a description of the data types of columns in the `SchemaRDD`.
+*Runtime checking*: Since `Pipeline`s can operate on datasets with varied types, they cannot use compile-time type checking. `Pipeline`s and `PipelineModel`s instead do runtime checking before actually running the `Pipeline`. This type checking is done using the dataset *schema*, a description of the data types of columns in the `DataFrame`.
## Parameters
@@ -171,12 +171,12 @@ import org.apache.spark.sql.{Row, SQLContext}
val conf = new SparkConf().setAppName("SimpleParamsExample")
val sc = new SparkContext(conf)
val sqlContext = new SQLContext(sc)
-import sqlContext._
+import sqlContext.implicits._
// Prepare training data.
// We use LabeledPoint, which is a case class. Spark SQL can convert RDDs of case classes
-// into SchemaRDDs, where it uses the case class metadata to infer the schema.
-val training = sparkContext.parallelize(Seq(
+// into DataFrames, where it uses the case class metadata to infer the schema.
+val training = sc.parallelize(Seq(
LabeledPoint(1.0, Vectors.dense(0.0, 1.1, 0.1)),
LabeledPoint(0.0, Vectors.dense(2.0, 1.0, -1.0)),
LabeledPoint(0.0, Vectors.dense(2.0, 1.3, 1.0)),
@@ -192,7 +192,7 @@ lr.setMaxIter(10)
.setRegParam(0.01)
// Learn a LogisticRegression model. This uses the parameters stored in lr.
-val model1 = lr.fit(training)
+val model1 = lr.fit(training.toDF)
// Since model1 is a Model (i.e., a Transformer produced by an Estimator),
// we can view the parameters it used during fit().
// This prints the parameter (name: value) pairs, where names are unique IDs for this
@@ -203,33 +203,35 @@ println("Model 1 was fit using parameters: " + model1.fittingParamMap)
// which supports several methods for specifying parameters.
val paramMap = ParamMap(lr.maxIter -> 20)
paramMap.put(lr.maxIter, 30) // Specify 1 Param. This overwrites the original maxIter.
-paramMap.put(lr.regParam -> 0.1, lr.threshold -> 0.5) // Specify multiple Params.
+paramMap.put(lr.regParam -> 0.1, lr.threshold -> 0.55) // Specify multiple Params.
// One can also combine ParamMaps.
-val paramMap2 = ParamMap(lr.scoreCol -> "probability") // Changes output column name.
+val paramMap2 = ParamMap(lr.probabilityCol -> "myProbability") // Change output column name
val paramMapCombined = paramMap ++ paramMap2
// Now learn a new model using the paramMapCombined parameters.
// paramMapCombined overrides all parameters set earlier via lr.set* methods.
-val model2 = lr.fit(training, paramMapCombined)
+val model2 = lr.fit(training.toDF, paramMapCombined)
println("Model 2 was fit using parameters: " + model2.fittingParamMap)
-// Prepare test documents.
-val test = sparkContext.parallelize(Seq(
+// Prepare test data.
+val test = sc.parallelize(Seq(
LabeledPoint(1.0, Vectors.dense(-1.0, 1.5, 1.3)),
LabeledPoint(0.0, Vectors.dense(3.0, 2.0, -0.1)),
LabeledPoint(1.0, Vectors.dense(0.0, 2.2, -1.5))))
-// Make predictions on test documents using the Transformer.transform() method.
+// Make predictions on test data using the Transformer.transform() method.
// LogisticRegression.transform will only use the 'features' column.
-// Note that model2.transform() outputs a 'probability' column instead of the usual 'score'
-// column since we renamed the lr.scoreCol parameter previously.
-model2.transform(test)
- .select('features, 'label, 'probability, 'prediction)
+// Note that model2.transform() outputs a 'myProbability' column instead of the usual
+// 'probability' column since we renamed the lr.probabilityCol parameter previously.
+model2.transform(test.toDF)
+ .select("features", "label", "myProbability", "prediction")
.collect()
- .foreach { case Row(features: Vector, label: Double, prob: Double, prediction: Double) =>
- println("(" + features + ", " + label + ") -> prob=" + prob + ", prediction=" + prediction)
+ .foreach { case Row(features: Vector, label: Double, prob: Vector, prediction: Double) =>
+ println("($features, $label) -> prob=$prob, prediction=$prediction")
}
+
+sc.stop()
{% endhighlight %}
</div>
@@ -244,23 +246,23 @@ import org.apache.spark.ml.param.ParamMap;
import org.apache.spark.ml.classification.LogisticRegression;
import org.apache.spark.mllib.linalg.Vectors;
import org.apache.spark.mllib.regression.LabeledPoint;
-import org.apache.spark.sql.api.java.JavaSQLContext;
-import org.apache.spark.sql.api.java.JavaSchemaRDD;
-import org.apache.spark.sql.api.java.Row;
+import org.apache.spark.sql.DataFrame;
+import org.apache.spark.sql.SQLContext;
+import org.apache.spark.sql.Row;
SparkConf conf = new SparkConf().setAppName("JavaSimpleParamsExample");
JavaSparkContext jsc = new JavaSparkContext(conf);
-JavaSQLContext jsql = new JavaSQLContext(jsc);
+SQLContext jsql = new SQLContext(jsc);
// Prepare training data.
-// We use LabeledPoint, which is a case class. Spark SQL can convert RDDs of case classes
-// into SchemaRDDs, where it uses the case class metadata to infer the schema.
+// We use LabeledPoint, which is a JavaBean. Spark SQL can convert RDDs of JavaBeans
+// into DataFrames, where it uses the bean metadata to infer the schema.
List<LabeledPoint> localTraining = Lists.newArrayList(
new LabeledPoint(1.0, Vectors.dense(0.0, 1.1, 0.1)),
new LabeledPoint(0.0, Vectors.dense(2.0, 1.0, -1.0)),
new LabeledPoint(0.0, Vectors.dense(2.0, 1.3, 1.0)),
new LabeledPoint(1.0, Vectors.dense(0.0, 1.2, -0.5)));
-JavaSchemaRDD training = jsql.createDataFrame(jsc.parallelize(localTraining), LabeledPoint.class);
+DataFrame training = jsql.createDataFrame(jsc.parallelize(localTraining), LabeledPoint.class);
// Create a LogisticRegression instance. This instance is an Estimator.
LogisticRegression lr = new LogisticRegression();
@@ -281,13 +283,13 @@ System.out.println("Model 1 was fit using parameters: " + model1.fittingParamMap
// We may alternatively specify parameters using a ParamMap.
ParamMap paramMap = new ParamMap();
-paramMap.put(lr.maxIter(), 20); // Specify 1 Param.
+paramMap.put(lr.maxIter().w(20)); // Specify 1 Param.
paramMap.put(lr.maxIter(), 30); // This overwrites the original maxIter.
-paramMap.put(lr.regParam(), 0.1);
+paramMap.put(lr.regParam().w(0.1), lr.threshold().w(0.55)); // Specify multiple Params.
// One can also combine ParamMaps.
ParamMap paramMap2 = new ParamMap();
-paramMap2.put(lr.scoreCol(), "probability"); // Changes output column name.
+paramMap2.put(lr.probabilityCol().w("myProbability")); // Change output column name
ParamMap paramMapCombined = paramMap.$plus$plus(paramMap2);
// Now learn a new model using the paramMapCombined parameters.
@@ -300,19 +302,19 @@ List<LabeledPoint> localTest = Lists.newArrayList(
new LabeledPoint(1.0, Vectors.dense(-1.0, 1.5, 1.3)),
new LabeledPoint(0.0, Vectors.dense(3.0, 2.0, -0.1)),
new LabeledPoint(1.0, Vectors.dense(0.0, 2.2, -1.5)));
-JavaSchemaRDD test = jsql.createDataFrame(jsc.parallelize(localTest), LabeledPoint.class);
+DataFrame test = jsql.createDataFrame(jsc.parallelize(localTest), LabeledPoint.class);
// Make predictions on test documents using the Transformer.transform() method.
// LogisticRegression.transform will only use the 'features' column.
-// Note that model2.transform() outputs a 'probability' column instead of the usual 'score'
-// column since we renamed the lr.scoreCol parameter previously.
-model2.transform(test).registerAsTable("results");
-JavaSchemaRDD results =
- jsql.sql("SELECT features, label, probability, prediction FROM results");
-for (Row r: results.collect()) {
+// Note that model2.transform() outputs a 'myProbability' column instead of the usual
+// 'probability' column since we renamed the lr.probabilityCol parameter previously.
+DataFrame results = model2.transform(test);
+for (Row r: results.select("features", "label", "myProbability", "prediction").collect()) {
System.out.println("(" + r.get(0) + ", " + r.get(1) + ") -> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
+
+jsc.stop();
{% endhighlight %}
</div>
@@ -330,6 +332,7 @@ import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.ml.Pipeline
import org.apache.spark.ml.classification.LogisticRegression
import org.apache.spark.ml.feature.{HashingTF, Tokenizer}
+import org.apache.spark.mllib.linalg.Vector
import org.apache.spark.sql.{Row, SQLContext}
// Labeled and unlabeled instance types.
@@ -337,14 +340,14 @@ import org.apache.spark.sql.{Row, SQLContext}
case class LabeledDocument(id: Long, text: String, label: Double)
case class Document(id: Long, text: String)
-// Set up contexts. Import implicit conversions to SchemaRDD from sqlContext.
+// Set up contexts. Import implicit conversions to DataFrame from sqlContext.
val conf = new SparkConf().setAppName("SimpleTextClassificationPipeline")
val sc = new SparkContext(conf)
val sqlContext = new SQLContext(sc)
-import sqlContext._
+import sqlContext.implicits._
// Prepare training documents, which are labeled.
-val training = sparkContext.parallelize(Seq(
+val training = sc.parallelize(Seq(
LabeledDocument(0L, "a b c d e spark", 1.0),
LabeledDocument(1L, "b d", 0.0),
LabeledDocument(2L, "spark f g h", 1.0),
@@ -365,30 +368,32 @@ val pipeline = new Pipeline()
.setStages(Array(tokenizer, hashingTF, lr))
// Fit the pipeline to training documents.
-val model = pipeline.fit(training)
+val model = pipeline.fit(training.toDF)
// Prepare test documents, which are unlabeled.
-val test = sparkContext.parallelize(Seq(
+val test = sc.parallelize(Seq(
Document(4L, "spark i j k"),
Document(5L, "l m n"),
Document(6L, "mapreduce spark"),
Document(7L, "apache hadoop")))
// Make predictions on test documents.
-model.transform(test)
- .select('id, 'text, 'score, 'prediction)
+model.transform(test.toDF)
+ .select("id", "text", "probability", "prediction")
.collect()
- .foreach { case Row(id: Long, text: String, score: Double, prediction: Double) =>
- println("(" + id + ", " + text + ") --> score=" + score + ", prediction=" + prediction)
+ .foreach { case Row(id: Long, text: String, prob: Vector, prediction: Double) =>
+ println("($id, $text) --> prob=$prob, prediction=$prediction")
}
+
+sc.stop()
{% endhighlight %}
</div>
<div data-lang="java">
{% highlight java %}
-import java.io.Serializable;
import java.util.List;
import com.google.common.collect.Lists;
+import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext;
import org.apache.spark.ml.Pipeline;
import org.apache.spark.ml.PipelineModel;
@@ -396,10 +401,9 @@ import org.apache.spark.ml.PipelineStage;
import org.apache.spark.ml.classification.LogisticRegression;
import org.apache.spark.ml.feature.HashingTF;
import org.apache.spark.ml.feature.Tokenizer;
-import org.apache.spark.sql.api.java.JavaSQLContext;
-import org.apache.spark.sql.api.java.JavaSchemaRDD;
-import org.apache.spark.sql.api.java.Row;
-import org.apache.spark.SparkConf;
+import org.apache.spark.sql.DataFrame;
+import org.apache.spark.sql.Row;
+import org.apache.spark.sql.SQLContext;
// Labeled and unlabeled instance types.
// Spark SQL can infer schema from Java Beans.
@@ -434,7 +438,7 @@ public class LabeledDocument extends Document implements Serializable {
// Set up contexts.
SparkConf conf = new SparkConf().setAppName("JavaSimpleTextClassificationPipeline");
JavaSparkContext jsc = new JavaSparkContext(conf);
-JavaSQLContext jsql = new JavaSQLContext(jsc);
+SQLContext jsql = new SQLContext(jsc);
// Prepare training documents, which are labeled.
List<LabeledDocument> localTraining = Lists.newArrayList(
@@ -442,8 +446,7 @@ List<LabeledDocument> localTraining = Lists.newArrayList(
new LabeledDocument(1L, "b d", 0.0),
new LabeledDocument(2L, "spark f g h", 1.0),
new LabeledDocument(3L, "hadoop mapreduce", 0.0));
-JavaSchemaRDD training =
- jsql.createDataFrame(jsc.parallelize(localTraining), LabeledDocument.class);
+DataFrame training = jsql.createDataFrame(jsc.parallelize(localTraining), LabeledDocument.class);
// Configure an ML pipeline, which consists of three stages: tokenizer, hashingTF, and lr.
Tokenizer tokenizer = new Tokenizer()
@@ -468,16 +471,62 @@ List<Document> localTest = Lists.newArrayList(
new Document(5L, "l m n"),
new Document(6L, "mapreduce spark"),
new Document(7L, "apache hadoop"));
-JavaSchemaRDD test =
- jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
+DataFrame test = jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
// Make predictions on test documents.
-model.transform(test).registerAsTable("prediction");
-JavaSchemaRDD predictions = jsql.sql("SELECT id, text, score, prediction FROM prediction");
-for (Row r: predictions.collect()) {
- System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> score=" + r.get(2)
+DataFrame predictions = model.transform(test);
+for (Row r: predictions.select("id", "text", "probability", "prediction").collect()) {
+ System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
+
+jsc.stop();
+{% endhighlight %}
+</div>
+
+<div data-lang="python">
+{% highlight python %}
+from pyspark import SparkContext
+from pyspark.ml import Pipeline
+from pyspark.ml.classification import LogisticRegression
+from pyspark.ml.feature import HashingTF, Tokenizer
+from pyspark.sql import Row, SQLContext
+
+sc = SparkContext(appName="SimpleTextClassificationPipeline")
+sqlCtx = SQLContext(sc)
+
+# Prepare training documents, which are labeled.
+LabeledDocument = Row("id", "text", "label")
+training = sc.parallelize([(0L, "a b c d e spark", 1.0),
+ (1L, "b d", 0.0),
+ (2L, "spark f g h", 1.0),
+ (3L, "hadoop mapreduce", 0.0)]) \
+ .map(lambda x: LabeledDocument(*x)).toDF()
+
+# Configure an ML pipeline, which consists of tree stages: tokenizer, hashingTF, and lr.
+tokenizer = Tokenizer(inputCol="text", outputCol="words")
+hashingTF = HashingTF(inputCol=tokenizer.getOutputCol(), outputCol="features")
+lr = LogisticRegression(maxIter=10, regParam=0.01)
+pipeline = Pipeline(stages=[tokenizer, hashingTF, lr])
+
+# Fit the pipeline to training documents.
+model = pipeline.fit(training)
+
+# Prepare test documents, which are unlabeled.
+Document = Row("id", "text")
+test = sc.parallelize([(4L, "spark i j k"),
+ (5L, "l m n"),
+ (6L, "mapreduce spark"),
+ (7L, "apache hadoop")]) \
+ .map(lambda x: Document(*x)).toDF()
+
+# Make predictions on test documents and print columns of interest.
+prediction = model.transform(test)
+selected = prediction.select("id", "text", "prediction")
+for row in selected.collect():
+ print row
+
+sc.stop()
{% endhighlight %}
</div>
@@ -508,21 +557,21 @@ However, it is also a well-established method for choosing parameters which is m
<div data-lang="scala">
{% highlight scala %}
import org.apache.spark.{SparkConf, SparkContext}
-import org.apache.spark.SparkContext._
import org.apache.spark.ml.Pipeline
import org.apache.spark.ml.classification.LogisticRegression
import org.apache.spark.ml.evaluation.BinaryClassificationEvaluator
import org.apache.spark.ml.feature.{HashingTF, Tokenizer}
import org.apache.spark.ml.tuning.{ParamGridBuilder, CrossValidator}
+import org.apache.spark.mllib.linalg.Vector
import org.apache.spark.sql.{Row, SQLContext}
val conf = new SparkConf().setAppName("CrossValidatorExample")
val sc = new SparkContext(conf)
val sqlContext = new SQLContext(sc)
-import sqlContext._
+import sqlContext.implicits._
// Prepare training documents, which are labeled.
-val training = sparkContext.parallelize(Seq(
+val training = sc.parallelize(Seq(
LabeledDocument(0L, "a b c d e spark", 1.0),
LabeledDocument(1L, "b d", 0.0),
LabeledDocument(2L, "spark f g h", 1.0),
@@ -565,24 +614,24 @@ crossval.setEstimatorParamMaps(paramGrid)
crossval.setNumFolds(2) // Use 3+ in practice
// Run cross-validation, and choose the best set of parameters.
-val cvModel = crossval.fit(training)
-// Get the best LogisticRegression model (with the best set of parameters from paramGrid).
-val lrModel = cvModel.bestModel
+val cvModel = crossval.fit(training.toDF)
// Prepare test documents, which are unlabeled.
-val test = sparkContext.parallelize(Seq(
+val test = sc.parallelize(Seq(
Document(4L, "spark i j k"),
Document(5L, "l m n"),
Document(6L, "mapreduce spark"),
Document(7L, "apache hadoop")))
// Make predictions on test documents. cvModel uses the best model found (lrModel).
-cvModel.transform(test)
- .select('id, 'text, 'score, 'prediction)
+cvModel.transform(test.toDF)
+ .select("id", "text", "probability", "prediction")
.collect()
- .foreach { case Row(id: Long, text: String, score: Double, prediction: Double) =>
- println("(" + id + ", " + text + ") --> score=" + score + ", prediction=" + prediction)
+ .foreach { case Row(id: Long, text: String, prob: Vector, prediction: Double) =>
+ println(s"($id, $text) --> prob=$prob, prediction=$prediction")
}
+
+sc.stop()
{% endhighlight %}
</div>
@@ -592,7 +641,6 @@ import java.util.List;
import com.google.common.collect.Lists;
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.JavaSparkContext;
-import org.apache.spark.ml.Model;
import org.apache.spark.ml.Pipeline;
import org.apache.spark.ml.PipelineStage;
import org.apache.spark.ml.classification.LogisticRegression;
@@ -603,13 +651,13 @@ import org.apache.spark.ml.param.ParamMap;
import org.apache.spark.ml.tuning.CrossValidator;
import org.apache.spark.ml.tuning.CrossValidatorModel;
import org.apache.spark.ml.tuning.ParamGridBuilder;
-import org.apache.spark.sql.api.java.JavaSQLContext;
-import org.apache.spark.sql.api.java.JavaSchemaRDD;
-import org.apache.spark.sql.api.java.Row;
+import org.apache.spark.sql.DataFrame;
+import org.apache.spark.sql.Row;
+import org.apache.spark.sql.SQLContext;
SparkConf conf = new SparkConf().setAppName("JavaCrossValidatorExample");
JavaSparkContext jsc = new JavaSparkContext(conf);
-JavaSQLContext jsql = new JavaSQLContext(jsc);
+SQLContext jsql = new SQLContext(jsc);
// Prepare training documents, which are labeled.
List<LabeledDocument> localTraining = Lists.newArrayList(
@@ -625,8 +673,7 @@ List<LabeledDocument> localTraining = Lists.newArrayList(
new LabeledDocument(9L, "a e c l", 0.0),
new LabeledDocument(10L, "spark compile", 1.0),
new LabeledDocument(11L, "hadoop software", 0.0));
-JavaSchemaRDD training =
- jsql.createDataFrame(jsc.parallelize(localTraining), LabeledDocument.class);
+DataFrame training = jsql.createDataFrame(jsc.parallelize(localTraining), LabeledDocument.class);
// Configure an ML pipeline, which consists of three stages: tokenizer, hashingTF, and lr.
Tokenizer tokenizer = new Tokenizer()
@@ -660,8 +707,6 @@ crossval.setNumFolds(2); // Use 3+ in practice
// Run cross-validation, and choose the best set of parameters.
CrossValidatorModel cvModel = crossval.fit(training);
-// Get the best LogisticRegression model (with the best set of parameters from paramGrid).
-Model lrModel = cvModel.bestModel();
// Prepare test documents, which are unlabeled.
List<Document> localTest = Lists.newArrayList(
@@ -669,15 +714,16 @@ List<Document> localTest = Lists.newArrayList(
new Document(5L, "l m n"),
new Document(6L, "mapreduce spark"),
new Document(7L, "apache hadoop"));
-JavaSchemaRDD test = jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
+DataFrame test = jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
// Make predictions on test documents. cvModel uses the best model found (lrModel).
-cvModel.transform(test).registerAsTable("prediction");
-JavaSchemaRDD predictions = jsql.sql("SELECT id, text, score, prediction FROM prediction");
-for (Row r: predictions.collect()) {
- System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> score=" + r.get(2)
+DataFrame predictions = cvModel.transform(test);
+for (Row r: predictions.select("id", "text", "probability", "prediction").collect()) {
+ System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
+
+jsc.stop();
{% endhighlight %}
</div>
@@ -686,6 +732,21 @@ for (Row r: predictions.collect()) {
# Dependencies
Spark ML currently depends on MLlib and has the same dependencies.
-Please see the [MLlib Dependencies guide](mllib-guide.html#Dependencies) for more info.
+Please see the [MLlib Dependencies guide](mllib-guide.html#dependencies) for more info.
Spark ML also depends upon Spark SQL, but the relevant parts of Spark SQL do not bring additional dependencies.
+
+# Migration Guide
+
+## From 1.2 to 1.3
+
+The main API changes are from Spark SQL. We list the most important changes here:
+
+* The old [SchemaRDD](http://spark.apache.org/docs/1.2.1/api/scala/index.html#org.apache.spark.sql.SchemaRDD) has been replaced with [DataFrame](api/scala/index.html#org.apache.spark.sql.DataFrame) with a somewhat modified API. All algorithms in Spark ML which used to use SchemaRDD now use DataFrame.
+* In Spark 1.2, we used implicit conversions from `RDD`s of `LabeledPoint` into `SchemaRDD`s by calling `import sqlContext._` where `sqlContext` was an instance of `SQLContext`. These implicits have been moved, so we now call `import sqlContext.implicits._`.
+* Java APIs for SQL have also changed accordingly. Please see the examples above and the [Spark SQL Programming Guide](sql-programming-guide.html) for details.
+
+Other changes were in `LogisticRegression`:
+
+* The `scoreCol` output column (with default value "score") was renamed to be `probabilityCol` (with default value "probability"). The type was originally `Double` (for the probability of class 1.0), but it is now `Vector` (for the probability of each class, to support multiclass classification in the future).
+* In Spark 1.2, `LogisticRegressionModel` did not include an intercept. In Spark 1.3, it includes an intercept; however, it will always be 0.0 since it uses the default settings for [spark.mllib.LogisticRegressionWithLBFGS](api/scala/index.html#org.apache.spark.mllib.classification.LogisticRegressionWithLBFGS). The option to use an intercept will be added in the future.
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/docs/mllib-guide.md
----------------------------------------------------------------------
diff --git a/docs/mllib-guide.md b/docs/mllib-guide.md
index 0ca51f9..4c7a7d9 100644
--- a/docs/mllib-guide.md
+++ b/docs/mllib-guide.md
@@ -46,7 +46,7 @@ and the migration guide below will explain all changes between releases.
# spark.ml: high-level APIs for ML pipelines
-Spark 1.2 includes a new package called `spark.ml`, which aims to provide a uniform set of
+Spark 1.2 introduced a new package called `spark.ml`, which aims to provide a uniform set of
high-level APIs that help users create and tune practical machine learning pipelines.
It is currently an alpha component, and we would like to hear back from the community about
how it fits real-world use cases and how it could be improved.
@@ -92,125 +92,22 @@ version 1.4 or newer.
# Migration Guide
-## From 1.1 to 1.2
+For the `spark.ml` package, please see the [spark.ml Migration Guide](ml-guide.html#migration-guide).
-The only API changes in MLlib v1.2 are in
-[`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
-which continues to be an experimental API in MLlib 1.2:
+## From 1.2 to 1.3
-1. *(Breaking change)* The Scala API for classification takes a named argument specifying the number
-of classes. In MLlib v1.1, this argument was called `numClasses` in Python and
-`numClassesForClassification` in Scala. In MLlib v1.2, the names are both set to `numClasses`.
-This `numClasses` parameter is specified either via
-[`Strategy`](api/scala/index.html#org.apache.spark.mllib.tree.configuration.Strategy)
-or via [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree)
-static `trainClassifier` and `trainRegressor` methods.
+In the `spark.mllib` package, there were several breaking changes. The first change (in `ALS`) is the only one in a component not marked as Alpha or Experimental.
-2. *(Breaking change)* The API for
-[`Node`](api/scala/index.html#org.apache.spark.mllib.tree.model.Node) has changed.
-This should generally not affect user code, unless the user manually constructs decision trees
-(instead of using the `trainClassifier` or `trainRegressor` methods).
-The tree `Node` now includes more information, including the probability of the predicted label
-(for classification).
+* *(Breaking change)* In [`ALS`](api/scala/index.html#org.apache.spark.mllib.recommendation.ALS), the extraneous method `solveLeastSquares` has been removed. The `DeveloperApi` method `analyzeBlocks` was also removed.
+* *(Breaking change)* [`StandardScalerModel`](api/scala/index.html#org.apache.spark.mllib.feature.StandardScalerModel) remains an Alpha component. In it, the `variance` method has been replaced with the `std` method. To compute the column variance values returned by the original `variance` method, simply square the standard deviation values returned by `std`.
+* *(Breaking change)* [`StreamingLinearRegressionWithSGD`](api/scala/index.html#org.apache.spark.mllib.regression.StreamingLinearRegressionWithSGD) remains an Experimental component. In it, there were two changes:
+ * The constructor taking arguments was removed in favor of a builder patten using the default constructor plus parameter setter methods.
+ * Variable `model` is no longer public.
+* *(Breaking change)* [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree) remains an Experimental component. In it and its associated classes, there were several changes:
+ * In `DecisionTree`, the deprecated class method `train` has been removed. (The object/static `train` methods remain.)
+ * In `Strategy`, the `checkpointDir` parameter has been removed. Checkpointing is still supported, but the checkpoint directory must be set before calling tree and tree ensemble training.
+* `PythonMLlibAPI` (the interface between Scala/Java and Python for MLlib) was a public API but is now private, declared `private[python]`. This was never meant for external use.
-3. Printing methods' output has changed. The `toString` (Scala/Java) and `__repr__` (Python) methods used to print the full model; they now print a summary. For the full model, use `toDebugString`.
+## Previous Spark Versions
-Examples in the Spark distribution and examples in the
-[Decision Trees Guide](mllib-decision-tree.html#examples) have been updated accordingly.
-
-## From 1.0 to 1.1
-
-The only API changes in MLlib v1.1 are in
-[`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
-which continues to be an experimental API in MLlib 1.1:
-
-1. *(Breaking change)* The meaning of tree depth has been changed by 1 in order to match
-the implementations of trees in
-[scikit-learn](http://scikit-learn.org/stable/modules/classes.html#module-sklearn.tree)
-and in [rpart](http://cran.r-project.org/web/packages/rpart/index.html).
-In MLlib v1.0, a depth-1 tree had 1 leaf node, and a depth-2 tree had 1 root node and 2 leaf nodes.
-In MLlib v1.1, a depth-0 tree has 1 leaf node, and a depth-1 tree has 1 root node and 2 leaf nodes.
-This depth is specified by the `maxDepth` parameter in
-[`Strategy`](api/scala/index.html#org.apache.spark.mllib.tree.configuration.Strategy)
-or via [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree)
-static `trainClassifier` and `trainRegressor` methods.
-
-2. *(Non-breaking change)* We recommend using the newly added `trainClassifier` and `trainRegressor`
-methods to build a [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
-rather than using the old parameter class `Strategy`. These new training methods explicitly
-separate classification and regression, and they replace specialized parameter types with
-simple `String` types.
-
-Examples of the new, recommended `trainClassifier` and `trainRegressor` are given in the
-[Decision Trees Guide](mllib-decision-tree.html#examples).
-
-## From 0.9 to 1.0
-
-In MLlib v1.0, we support both dense and sparse input in a unified way, which introduces a few
-breaking changes. If your data is sparse, please store it in a sparse format instead of dense to
-take advantage of sparsity in both storage and computation. Details are described below.
-
-<div class="codetabs">
-<div data-lang="scala" markdown="1">
-
-We used to represent a feature vector by `Array[Double]`, which is replaced by
-[`Vector`](api/scala/index.html#org.apache.spark.mllib.linalg.Vector) in v1.0. Algorithms that used
-to accept `RDD[Array[Double]]` now take
-`RDD[Vector]`. [`LabeledPoint`](api/scala/index.html#org.apache.spark.mllib.regression.LabeledPoint)
-is now a wrapper of `(Double, Vector)` instead of `(Double, Array[Double])`. Converting
-`Array[Double]` to `Vector` is straightforward:
-
-{% highlight scala %}
-import org.apache.spark.mllib.linalg.{Vector, Vectors}
-
-val array: Array[Double] = ... // a double array
-val vector: Vector = Vectors.dense(array) // a dense vector
-{% endhighlight %}
-
-[`Vectors`](api/scala/index.html#org.apache.spark.mllib.linalg.Vectors$) provides factory methods to create sparse vectors.
-
-*Note*: Scala imports `scala.collection.immutable.Vector` by default, so you have to import `org.apache.spark.mllib.linalg.Vector` explicitly to use MLlib's `Vector`.
-
-</div>
-
-<div data-lang="java" markdown="1">
-
-We used to represent a feature vector by `double[]`, which is replaced by
-[`Vector`](api/java/index.html?org/apache/spark/mllib/linalg/Vector.html) in v1.0. Algorithms that used
-to accept `RDD<double[]>` now take
-`RDD<Vector>`. [`LabeledPoint`](api/java/index.html?org/apache/spark/mllib/regression/LabeledPoint.html)
-is now a wrapper of `(double, Vector)` instead of `(double, double[])`. Converting `double[]` to
-`Vector` is straightforward:
-
-{% highlight java %}
-import org.apache.spark.mllib.linalg.Vector;
-import org.apache.spark.mllib.linalg.Vectors;
-
-double[] array = ... // a double array
-Vector vector = Vectors.dense(array); // a dense vector
-{% endhighlight %}
-
-[`Vectors`](api/scala/index.html#org.apache.spark.mllib.linalg.Vectors$) provides factory methods to
-create sparse vectors.
-
-</div>
-
-<div data-lang="python" markdown="1">
-
-We used to represent a labeled feature vector in a NumPy array, where the first entry corresponds to
-the label and the rest are features. This representation is replaced by class
-[`LabeledPoint`](api/python/pyspark.mllib.regression.LabeledPoint-class.html), which takes both
-dense and sparse feature vectors.
-
-{% highlight python %}
-from pyspark.mllib.linalg import SparseVector
-from pyspark.mllib.regression import LabeledPoint
-
-# Create a labeled point with a positive label and a dense feature vector.
-pos = LabeledPoint(1.0, [1.0, 0.0, 3.0])
-
-# Create a labeled point with a negative label and a sparse feature vector.
-neg = LabeledPoint(0.0, SparseVector(3, [0, 2], [1.0, 3.0]))
-{% endhighlight %}
-</div>
-</div>
+Earlier migration guides are archived [on this page](mllib-migration-guides.html).
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/docs/mllib-migration-guides.md
----------------------------------------------------------------------
diff --git a/docs/mllib-migration-guides.md b/docs/mllib-migration-guides.md
new file mode 100644
index 0000000..4de2d94
--- /dev/null
+++ b/docs/mllib-migration-guides.md
@@ -0,0 +1,67 @@
+---
+layout: global
+title: Old Migration Guides - MLlib
+displayTitle: <a href="mllib-guide.html">MLlib</a> - Old Migration Guides
+description: MLlib migration guides from before Spark SPARK_VERSION_SHORT
+---
+
+The migration guide for the current Spark version is kept on the [MLlib Programming Guide main page](mllib-guide.html#migration-guide).
+
+## From 1.1 to 1.2
+
+The only API changes in MLlib v1.2 are in
+[`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
+which continues to be an experimental API in MLlib 1.2:
+
+1. *(Breaking change)* The Scala API for classification takes a named argument specifying the number
+of classes. In MLlib v1.1, this argument was called `numClasses` in Python and
+`numClassesForClassification` in Scala. In MLlib v1.2, the names are both set to `numClasses`.
+This `numClasses` parameter is specified either via
+[`Strategy`](api/scala/index.html#org.apache.spark.mllib.tree.configuration.Strategy)
+or via [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree)
+static `trainClassifier` and `trainRegressor` methods.
+
+2. *(Breaking change)* The API for
+[`Node`](api/scala/index.html#org.apache.spark.mllib.tree.model.Node) has changed.
+This should generally not affect user code, unless the user manually constructs decision trees
+(instead of using the `trainClassifier` or `trainRegressor` methods).
+The tree `Node` now includes more information, including the probability of the predicted label
+(for classification).
+
+3. Printing methods' output has changed. The `toString` (Scala/Java) and `__repr__` (Python) methods used to print the full model; they now print a summary. For the full model, use `toDebugString`.
+
+Examples in the Spark distribution and examples in the
+[Decision Trees Guide](mllib-decision-tree.html#examples) have been updated accordingly.
+
+## From 1.0 to 1.1
+
+The only API changes in MLlib v1.1 are in
+[`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
+which continues to be an experimental API in MLlib 1.1:
+
+1. *(Breaking change)* The meaning of tree depth has been changed by 1 in order to match
+the implementations of trees in
+[scikit-learn](http://scikit-learn.org/stable/modules/classes.html#module-sklearn.tree)
+and in [rpart](http://cran.r-project.org/web/packages/rpart/index.html).
+In MLlib v1.0, a depth-1 tree had 1 leaf node, and a depth-2 tree had 1 root node and 2 leaf nodes.
+In MLlib v1.1, a depth-0 tree has 1 leaf node, and a depth-1 tree has 1 root node and 2 leaf nodes.
+This depth is specified by the `maxDepth` parameter in
+[`Strategy`](api/scala/index.html#org.apache.spark.mllib.tree.configuration.Strategy)
+or via [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree)
+static `trainClassifier` and `trainRegressor` methods.
+
+2. *(Non-breaking change)* We recommend using the newly added `trainClassifier` and `trainRegressor`
+methods to build a [`DecisionTree`](api/scala/index.html#org.apache.spark.mllib.tree.DecisionTree),
+rather than using the old parameter class `Strategy`. These new training methods explicitly
+separate classification and regression, and they replace specialized parameter types with
+simple `String` types.
+
+Examples of the new, recommended `trainClassifier` and `trainRegressor` are given in the
+[Decision Trees Guide](mllib-decision-tree.html#examples).
+
+## From 0.9 to 1.0
+
+In MLlib v1.0, we support both dense and sparse input in a unified way, which introduces a few
+breaking changes. If your data is sparse, please store it in a sparse format instead of dense to
+take advantage of sparsity in both storage and computation. Details are described below.
+
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/java/org/apache/spark/examples/ml/JavaCrossValidatorExample.java
----------------------------------------------------------------------
diff --git a/examples/src/main/java/org/apache/spark/examples/ml/JavaCrossValidatorExample.java b/examples/src/main/java/org/apache/spark/examples/ml/JavaCrossValidatorExample.java
index 5d8c5d0..9bbc14e 100644
--- a/examples/src/main/java/org/apache/spark/examples/ml/JavaCrossValidatorExample.java
+++ b/examples/src/main/java/org/apache/spark/examples/ml/JavaCrossValidatorExample.java
@@ -34,8 +34,8 @@ import org.apache.spark.ml.tuning.CrossValidator;
import org.apache.spark.ml.tuning.CrossValidatorModel;
import org.apache.spark.ml.tuning.ParamGridBuilder;
import org.apache.spark.sql.DataFrame;
-import org.apache.spark.sql.SQLContext;
import org.apache.spark.sql.Row;
+import org.apache.spark.sql.SQLContext;
/**
* A simple example demonstrating model selection using CrossValidator.
@@ -115,9 +115,8 @@ public class JavaCrossValidatorExample {
DataFrame test = jsql.createDataFrame(jsc.parallelize(localTest), Document.class);
// Make predictions on test documents. cvModel uses the best model found (lrModel).
- cvModel.transform(test).registerTempTable("prediction");
- DataFrame predictions = jsql.sql("SELECT id, text, probability, prediction FROM prediction");
- for (Row r: predictions.collect()) {
+ DataFrame predictions = cvModel.transform(test);
+ for (Row r: predictions.select("id", "text", "probability", "prediction").collect()) {
System.out.println("(" + r.get(0) + ", " + r.get(1) + ") --> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
http://git-wip-us.apache.org/repos/asf/spark/blob/4a17eedb/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleParamsExample.java
----------------------------------------------------------------------
diff --git a/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleParamsExample.java b/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleParamsExample.java
index 4c4d532..4e02acc 100644
--- a/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleParamsExample.java
+++ b/examples/src/main/java/org/apache/spark/examples/ml/JavaSimpleParamsExample.java
@@ -29,8 +29,8 @@ import org.apache.spark.ml.classification.LogisticRegression;
import org.apache.spark.mllib.linalg.Vectors;
import org.apache.spark.mllib.regression.LabeledPoint;
import org.apache.spark.sql.DataFrame;
-import org.apache.spark.sql.SQLContext;
import org.apache.spark.sql.Row;
+import org.apache.spark.sql.SQLContext;
/**
* A simple example demonstrating ways to specify parameters for Estimators and Transformers.
@@ -100,10 +100,8 @@ public class JavaSimpleParamsExample {
// LogisticRegression.transform will only use the 'features' column.
// Note that model2.transform() outputs a 'myProbability' column instead of the usual
// 'probability' column since we renamed the lr.probabilityCol parameter previously.
- model2.transform(test).registerTempTable("results");
- DataFrame results =
- jsql.sql("SELECT features, label, myProbability, prediction FROM results");
- for (Row r: results.collect()) {
+ DataFrame results = model2.transform(test);
+ for (Row r: results.select("features", "label", "myProbability", "prediction").collect()) {
System.out.println("(" + r.get(0) + ", " + r.get(1) + ") -> prob=" + r.get(2)
+ ", prediction=" + r.get(3));
}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@spark.apache.org
For additional commands, e-mail: commits-help@spark.apache.org