You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* From the command line, invoke a script with gradlew using the appropriate run<_ScriptName_> task.
46
-
(Hint: `gradlew :HousePrices:tasks --group="Script"` will show you available task names.)
51
+
(Hint: `gradlew :HousePrices:tasks --group="Application"` will show you available task names.)
47
52
* If the example has @Grab statements commented out at the top, you can cut and paste the examples into the groovyConsole
48
53
and uncomment the grab statements. Make sure to cut and paste any helper classes too if appropriate.
49
54
@@ -67,18 +72,22 @@ Some regression algorithm variants like stochastic gradient descent are amenable
67
72
And some frameworks support such algorithms. The following subprojects highlight frameworks
68
73
with special support for scaling linear regression:
69
74
70
-
* The [HousePricesIgnite](subprojects/HousePricesIgnite/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Ignite. It has been tested on JDK8, JDK11 and JDK17.
75
+
* The [HousePricesIgnite](../HousePricesIgnite/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Ignite. It has been tested on JDK8, JDK11 and JDK17.
71
76
72
-
* The [HousePricesSpark](subprojects/HousePricesSpark/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Spark. It has been tested on JDK8 and JDK11. The current Spark versions are not compatible with JDK17.
77
+
* The [HousePricesSpark](../HousePricesSpark/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Spark. It has been tested on JDK8 and JDK11. The current Spark versions are not compatible with JDK17.
73
78
74
79
If you find that your algorithm isn't directly amenable to scaling
75
80
you can often tweak it or apply it in some fashion to ensure certain
76
81
constraints hold. This can enable you to still scale up.
77
82
The following subprojects highlight tweaking linear regression
78
83
for scaling purposes:
79
84
80
-
* The [HousePricesBeam](subprojects/HousePricesBeam/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Beam. It has been tested on JDK8, JDK11 and JDK17.
85
+
* The [HousePricesBeam](../HousePricesBeam/src/main/groovy) subproject which illustrates scaling up to a cluster using Apache Beam. It has been tested on JDK8, JDK11 and JDK17.
86
+
87
+
* The [HousePricesGPars](../HousePricesGPars/src/main/groovy) subproject which illustrates scaling up concurrently using GPars. It has been tested on JDK8, JDK11 and JDK17.
81
88
82
-
* The [HousePricesGPars](subprojects/HousePricesGPars/src/main/groovy) subproject which illustrates scaling up concurrently using GPars. It has been tested on JDK8, JDK11 and JDK17.
89
+
When collecting data, it can be useful to use special purpose integration technologies:
83
90
84
-
When collecting data, it can be useful to use integration technologies like Apache Camel. The `ExploreOutlier_ApacheCamelCSV` example illustrates gathering data and finding outliers using this technology. This example works on JDK8, JDK11 and JDK17. An older version of Apache Camel is used on JDK8.
91
+
* The [HousePricesCamel](../HousePricesCamel/src/main/groovy) subproject
92
+
illustrates gathering data and finding outliers using Apache Camel.
0 commit comments