From e2ec32dab8530aa21ec95a27d60b1c22f3d1a18c Mon Sep 17 00:00:00 2001
From: hyukjinkwon <gurwls223@gmail.com>
Date: Thu, 19 May 2016 08:52:41 +0200
Subject: [PATCH] [SPARK-15031][EXAMPLES][FOLLOW-UP] Make Python param example
working with SparkSession
## What changes were proposed in this pull request?
It seems most of Python examples were changed to use SparkSession by https://github.com/apache/spark/pull/12809. This PR said both examples below:
- `simple_params_example.py`
- `aft_survival_regression.py`
are not changed because it dose not work. It seems `aft_survival_regression.py` is changed by https://github.com/apache/spark/pull/13050 but `simple_params_example.py` is not yet.
This PR corrects the example and make this use SparkSession.
In more detail, it seems `threshold` is replaced to `thresholds` here and there by https://github.com/apache/spark/commit/5a23213c148bfe362514f9c71f5273ebda0a848a. However, when it calls `lr.fit(training, paramMap)` this overwrites the values. So, `threshold` was 5 and `thresholds` becomes 5.5 (by `1 / (1 + thresholds(0) / thresholds(1)`).
According to the comment below. this is not allowed, https://github.com/apache/spark/blob/354f8f11bd4b20fa99bd67a98da3525fd3d75c81/mllib/src/main/scala/org/apache/spark/ml/classification/LogisticRegression.scala#L58-L61.
So, in this PR, it sets the equivalent value so that this does not throw an exception.
## How was this patch tested?
Manully (`mvn package -DskipTests && spark-submit simple_params_example.py`)
Author: hyukjinkwon <gurwls223@gmail.com>
Closes #13135 from HyukjinKwon/SPARK-15031.
---
.../examples/ml/JavaSimpleParamsExample.java | 2 +-
.../main/python/ml/simple_params_example.py | 24 +++++++++----------
.../examples/ml/SimpleParamsExample.scala | 2 +-
3 files changed, 13 insertions(+), 15 deletions(-)
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 ff1eb07dc6..ca80d0d8bb 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
@@ -77,7 +77,7 @@ public class JavaSimpleParamsExample {
ParamMap paramMap = new ParamMap();
paramMap.put(lr.maxIter().w(20)); // Specify 1 Param.
paramMap.put(lr.maxIter(), 30); // This overwrites the original maxIter.
- double[] thresholds = {0.45, 0.55};
+ double[] thresholds = {0.5, 0.5};
paramMap.put(lr.regParam().w(0.1), lr.thresholds().w(thresholds)); // Specify multiple Params.
// One can also combine ParamMaps.
diff --git a/examples/src/main/python/ml/simple_params_example.py b/examples/src/main/python/ml/simple_params_example.py
index 2d6d115d54..c57e59d01b 100644
--- a/examples/src/main/python/ml/simple_params_example.py
+++ b/examples/src/main/python/ml/simple_params_example.py
@@ -20,11 +20,10 @@ from __future__ import print_function
import pprint
import sys
-from pyspark import SparkContext
from pyspark.ml.classification import LogisticRegression
from pyspark.mllib.linalg import DenseVector
from pyspark.mllib.regression import LabeledPoint
-from pyspark.sql import SQLContext
+from pyspark.sql import SparkSession
"""
A simple example demonstrating ways to specify parameters for Estimators and Transformers.
@@ -33,21 +32,20 @@ Run with:
"""
if __name__ == "__main__":
- if len(sys.argv) > 1:
- print("Usage: simple_params_example", file=sys.stderr)
- exit(1)
- sc = SparkContext(appName="PythonSimpleParamsExample")
- sqlContext = SQLContext(sc)
+ spark = SparkSession \
+ .builder \
+ .appName("SimpleTextClassificationPipeline") \
+ .getOrCreate()
# prepare training data.
# We create an RDD of LabeledPoints and convert them into a DataFrame.
# A LabeledPoint is an Object with two fields named label and features
# and Spark SQL identifies these fields and creates the schema appropriately.
- training = sc.parallelize([
+ training = spark.createDataFrame([
LabeledPoint(1.0, DenseVector([0.0, 1.1, 0.1])),
LabeledPoint(0.0, DenseVector([2.0, 1.0, -1.0])),
LabeledPoint(0.0, DenseVector([2.0, 1.3, 1.0])),
- LabeledPoint(1.0, DenseVector([0.0, 1.2, -0.5]))]).toDF()
+ LabeledPoint(1.0, DenseVector([0.0, 1.2, -0.5]))])
# Create a LogisticRegression instance with maxIter = 10.
# This instance is an Estimator.
@@ -70,7 +68,7 @@ if __name__ == "__main__":
# We may alternatively specify parameters using a parameter map.
# paramMap overrides all lr parameters set earlier.
- paramMap = {lr.maxIter: 20, lr.thresholds: [0.45, 0.55], lr.probabilityCol: "myProbability"}
+ paramMap = {lr.maxIter: 20, lr.thresholds: [0.5, 0.5], lr.probabilityCol: "myProbability"}
# Now learn a new model using the new parameters.
model2 = lr.fit(training, paramMap)
@@ -78,10 +76,10 @@ if __name__ == "__main__":
pprint.pprint(model2.extractParamMap())
# prepare test data.
- test = sc.parallelize([
+ test = spark.createDataFrame([
LabeledPoint(1.0, DenseVector([-1.0, 1.5, 1.3])),
LabeledPoint(0.0, DenseVector([3.0, 2.0, -0.1])),
- LabeledPoint(0.0, DenseVector([0.0, 2.2, -1.5]))]).toDF()
+ LabeledPoint(0.0, DenseVector([0.0, 2.2, -1.5]))])
# Make predictions on test data using the Transformer.transform() method.
# LogisticRegressionModel.transform will only use the 'features' column.
@@ -95,4 +93,4 @@ if __name__ == "__main__":
print("features=%s,label=%s -> prob=%s, prediction=%s"
% (row.features, row.label, row.myProbability, row.prediction))
- sc.stop()
+ spark.stop()
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 3355c8ffa2..29f1f50960 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
@@ -70,7 +70,7 @@ object SimpleParamsExample {
// 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.thresholds -> Array(0.45, 0.55)) // Specify multiple Params.
+ paramMap.put(lr.regParam -> 0.1, lr.thresholds -> Array(0.5, 0.5)) // Specify multiple Params.
// One can also combine ParamMaps.
val paramMap2 = ParamMap(lr.probabilityCol -> "myProbability") // Change output column name
--
GitLab