Github user etrain commented on a diff in the pull request:
https://github.com/apache/spark/pull/3637#discussion_r22693073
--- Diff:
mllib/src/main/scala/org/apache/spark/ml/classification/Classifier.scala ---
@@ -0,0 +1,198 @@
+/*
+ * 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.classification
+
+import org.apache.spark.annotation.{DeveloperApi, AlphaComponent}
+import org.apache.spark.ml.impl.estimator.{PredictionModel, Predictor,
PredictorParams}
+import org.apache.spark.ml.param.{Params, ParamMap, HasRawPredictionCol}
+import org.apache.spark.mllib.linalg.{Vector, VectorUDT}
+import org.apache.spark.sql._
+import org.apache.spark.sql.catalyst.analysis.Star
+
+/**
+ * :: DeveloperApi ::
+ * Params for classification.
+ */
+@DeveloperApi
+trait ClassifierParams extends PredictorParams
+ with HasRawPredictionCol {
+
+ override protected def validateAndTransformSchema(
+ schema: StructType,
+ paramMap: ParamMap,
+ fitting: Boolean,
+ featuresDataType: DataType): StructType = {
+ val parentSchema = super.validateAndTransformSchema(schema, paramMap,
fitting, featuresDataType)
+ val map = this.paramMap ++ paramMap
+ addOutputColumn(parentSchema, map(rawPredictionCol), new VectorUDT)
+ }
+}
+
+/**
+ * :: AlphaComponent ::
+ * Single-label binary or multiclass classification.
+ * Classes are indexed {0, 1, ..., numClasses - 1}.
+ *
+ * @tparam FeaturesType Type of input features. E.g., [[Vector]]
+ * @tparam Learner Concrete Estimator type
+ * @tparam M Concrete Model type
+ */
+@AlphaComponent
+abstract class Classifier[
+ FeaturesType,
+ Learner <: Classifier[FeaturesType, Learner, M],
+ M <: ClassificationModel[FeaturesType, M]]
+ extends Predictor[FeaturesType, Learner, M]
+ with ClassifierParams {
+
+ def setRawPredictionCol(value: String): Learner =
+ set(rawPredictionCol, value).asInstanceOf[Learner]
+
+ // TODO: defaultEvaluator (follow-up PR)
+}
+
+/**
+ * :: AlphaComponent ::
+ * Model produced by a [[Classifier]].
+ * Classes are indexed {0, 1, ..., numClasses - 1}.
+ *
+ * @tparam FeaturesType Type of input features. E.g., [[Vector]]
+ * @tparam M Concrete Model type
+ */
+@AlphaComponent
+abstract class ClassificationModel[FeaturesType, M <:
ClassificationModel[FeaturesType, M]]
+ extends PredictionModel[FeaturesType, M] with ClassifierParams {
+
+ def setRawPredictionCol(value: String): M = set(rawPredictionCol,
value).asInstanceOf[M]
+
+ /** Number of classes (values which the label can take). */
+ def numClasses: Int
--- End diff --
How hard/weird would it be to make labels an Enumeration? This class could
be inferred from the training set at run-time or supplied by the user, and then
the user doesn't pass the number of classes to the model, but instead what the
set of labels actually are.
---
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]
