Change warning message text.

apache · Sep 25, 2014 · 7cca1dc · 7cca1dc
1 parent c77e939
commit 7cca1dc
Show file tree

Hide file tree

Showing 3 changed files with 12 additions and 6 deletions.
diff --git a/mllib/src/main/scala/org/apache/spark/mllib/clustering/KMeans.scala b/mllib/src/main/scala/org/apache/spark/mllib/clustering/KMeans.scala
@@ -121,7 +121,8 @@ class KMeans private (
 
     if (data.getStorageLevel == StorageLevel.NONE) {
       // Warn when running an iterative algorithm on uncached data. SPARK-1484
-      logWarning("KMeans.run called with uncached input data.")
+      logWarning("The input data is not directly cached, which may hurt performance if its"
+        + " parent RDDs are also uncached.")
     }
 
     // Compute squared norms and cache them.
@@ -134,7 +135,8 @@ class KMeans private (
     norms.unpersist()
 
     if (data.getStorageLevel == StorageLevel.NONE) {
-      logWarning("KMeans.run ran with uncached input data.")
+      logWarning("The input data was not directly cached, which may hurt performance if its"
+        + " parent RDDs are also uncached.")
     }
     model
   }

diff --git a/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala b/mllib/src/main/scala/org/apache/spark/mllib/linalg/distributed/RowMatrix.scala
@@ -234,7 +234,8 @@ class RowMatrix(
       case SVDMode.DistARPACK =>
         if (rows.getStorageLevel == StorageLevel.NONE) {
           // Warn when running an iterative algorithm on uncached data. SPARK-1484
-          logWarning("RowMatrix.computeSVD DistARPACK called with uncached input rows.")
+          logWarning("The input data is not directly cached, which may hurt performance if its"
+            + " parent RDDs are also uncached.")
         }
         require(k < n, s"k must be smaller than n in dist-eigs mode but got k=$k and n=$n.")
         EigenValueDecomposition.symmetricEigs(multiplyGramianMatrixBy, n, k, tol, maxIter)
@@ -262,7 +263,8 @@ class RowMatrix(
     }
 
     if (computeMode == SVDMode.DistARPACK && rows.getStorageLevel == StorageLevel.NONE) {
-      logWarning("RowMatrix.computeSVD DistARPACK ran with uncached input rows.")      
+      logWarning("The input data was not directly cached, which may hurt performance if its"
+        + " parent RDDs are also uncached.")
     }
 
     val s = Vectors.dense(Arrays.copyOfRange(sigmas.data, 0, sk))

diff --git a/mllib/src/main/scala/org/apache/spark/mllib/regression/GeneralizedLinearAlgorithm.scala b/mllib/src/main/scala/org/apache/spark/mllib/regression/GeneralizedLinearAlgorithm.scala
@@ -152,7 +152,8 @@ abstract class GeneralizedLinearAlgorithm[M <: GeneralizedLinearModel]
 
     if (input.getStorageLevel == StorageLevel.NONE) {
       // Warn when running an iterative algorithm on uncached data. SPARK-1484
-      logWarning("GeneralizedLinearAlgorithm.run called with uncached input data.")
+      logWarning("The input data is not directly cached, which may hurt performance if its"
+        + " parent RDDs are also uncached.")
     }
 
     // Check the data properties before running the optimizer
@@ -230,7 +231,8 @@ abstract class GeneralizedLinearAlgorithm[M <: GeneralizedLinearModel]
     }
 
     if (input.getStorageLevel == StorageLevel.NONE) {
-      logWarning("GeneralizedLinearAlgorithm.run ran with uncached input data.")
+      logWarning("The input data was not directly cached, which may hurt performance if its"
+        + " parent RDDs are also uncached.")
     }
 
     createModel(weights, intercept)