Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/3099#discussion_r20119991
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/tuning/CrossValidator.scala ---
@@ -0,0 +1,123 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.ml.tuning
+
+import com.github.fommil.netlib.F2jBLAS
+
+import org.apache.spark.Logging
+import org.apache.spark.ml._
+import org.apache.spark.ml.param.{IntParam, Param, ParamMap, Params}
+import org.apache.spark.mllib.util.MLUtils
+import org.apache.spark.sql.{SchemaRDD, StructType}
+
+/**
+ * Params for [[CrossValidator]] and [[CrossValidatorModel]].
+ */
+private[ml] trait CrossValidatorParams extends Params {
+ /** param for the estimator to be cross-validated */
+ val estimator: Param[Estimator[_]] = new Param(this, "estimator",
"estimator for selection")
+ def getEstimator: Estimator[_] = get(estimator)
+
+ /** param for estimator param maps */
+ val estimatorParamMaps: Param[Array[ParamMap]] =
+ new Param(this, "estimatorParamMaps", "param maps for the estimator")
+ def getEstimatorParamMaps: Array[ParamMap] = get(estimatorParamMaps)
+
+ /** param for the evaluator for selection */
+ val evaluator: Param[Evaluator] = new Param(this, "evaluator",
"evaluator for selection")
+ def getEvaluator: Evaluator = get(evaluator)
+
+ /** param for number of folds for cross validation */
+ val numFolds: IntParam =
+ new IntParam(this, "numFolds", "number of folds for cross validation",
Some(3))
+ def getNumFolds: Int = get(numFolds)
+}
+
+/**
+ * K-fold cross validation.
+ */
+class CrossValidator extends Estimator[CrossValidatorModel] with
CrossValidatorParams with Logging {
--- End diff --
Previously, I had supported having the estimator(s) param be passed as a
parameter, rather than to the constructor. I had been arguing for that since
it would allow you to do CV to choose between multiple estimator types, but I
had not realized that the current CV API does not allow that. (It only allows
1 estimator type.) If that is the plan, then I think CV should take the
estimator as a constructor parameter. That way, a user will be able to do the
following:
```
val lr = new LogisticRegression()
val cv = new CrossValidator(lr)
cv.params
```
With the current setup, "cv.params" will not be very helpful. But if CV
knows the estimator type, then it will be able to print out the set of relevant
parameters which the user can play with.
Alternatively, CV could take multiple estimators paired with ParamMaps,
which would make it easier to choose between model types.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
