Github user viirya commented on a diff in the pull request: https://github.com/apache/spark/pull/23100#discussion_r235760329 --- Diff: mllib/src/main/scala/org/apache/spark/ml/feature/OneHotEncoder.scala --- @@ -17,126 +17,512 @@ package org.apache.spark.ml.feature +import org.apache.hadoop.fs.Path + +import org.apache.spark.SparkException import org.apache.spark.annotation.Since -import org.apache.spark.ml.Transformer +import org.apache.spark.ml.{Estimator, Model} import org.apache.spark.ml.attribute._ import org.apache.spark.ml.linalg.Vectors import org.apache.spark.ml.param._ -import org.apache.spark.ml.param.shared.{HasInputCol, HasOutputCol} +import org.apache.spark.ml.param.shared.{HasHandleInvalid, HasInputCols, HasOutputCols} import org.apache.spark.ml.util._ import org.apache.spark.sql.{DataFrame, Dataset} -import org.apache.spark.sql.functions.{col, udf} -import org.apache.spark.sql.types.{DoubleType, NumericType, StructType} +import org.apache.spark.sql.expressions.UserDefinedFunction +import org.apache.spark.sql.functions.{col, lit, udf} +import org.apache.spark.sql.types.{DoubleType, StructField, StructType} + +/** Private trait for params and common methods for OneHotEncoder and OneHotEncoderModel */ +private[ml] trait OneHotEncoderBase extends Params with HasHandleInvalid + with HasInputCols with HasOutputCols { + + /** + * Param for how to handle invalid data during transform(). + * Options are 'keep' (invalid data presented as an extra categorical feature) or + * 'error' (throw an error). + * Note that this Param is only used during transform; during fitting, invalid data + * will result in an error. + * Default: "error" + * @group param + */ + @Since("2.3.0") + override val handleInvalid: Param[String] = new Param[String](this, "handleInvalid", + "How to handle invalid data during transform(). " + + "Options are 'keep' (invalid data presented as an extra categorical feature) " + + "or error (throw an error). Note that this Param is only used during transform; " + + "during fitting, invalid data will result in an error.", + ParamValidators.inArray(OneHotEncoder.supportedHandleInvalids)) + + setDefault(handleInvalid, OneHotEncoder.ERROR_INVALID) + + /** + * Whether to drop the last category in the encoded vector (default: true) + * @group param + */ + @Since("2.3.0") + final val dropLast: BooleanParam = + new BooleanParam(this, "dropLast", "whether to drop the last category") + setDefault(dropLast -> true) + + /** @group getParam */ + @Since("2.3.0") + def getDropLast: Boolean = $(dropLast) + + protected def validateAndTransformSchema( + schema: StructType, + dropLast: Boolean, + keepInvalid: Boolean): StructType = { + val inputColNames = $(inputCols) + val outputColNames = $(outputCols) + + require(inputColNames.length == outputColNames.length, + s"The number of input columns ${inputColNames.length} must be the same as the number of " + + s"output columns ${outputColNames.length}.") + + // Input columns must be NumericType. + inputColNames.foreach(SchemaUtils.checkNumericType(schema, _)) + + // Prepares output columns with proper attributes by examining input columns. + val inputFields = $(inputCols).map(schema(_)) + + val outputFields = inputFields.zip(outputColNames).map { case (inputField, outputColName) => + OneHotEncoderCommon.transformOutputColumnSchema( + inputField, outputColName, dropLast, keepInvalid) + } + outputFields.foldLeft(schema) { case (newSchema, outputField) => + SchemaUtils.appendColumn(newSchema, outputField) + } + } +} /** * A one-hot encoder that maps a column of category indices to a column of binary vectors, with * at most a single one-value per row that indicates the input category index. * For example with 5 categories, an input value of 2.0 would map to an output vector of * `[0.0, 0.0, 1.0, 0.0]`. - * The last category is not included by default (configurable via `OneHotEncoder!.dropLast` + * The last category is not included by default (configurable via `dropLast`), * because it makes the vector entries sum up to one, and hence linearly dependent. * So an input value of 4.0 maps to `[0.0, 0.0, 0.0, 0.0]`. * * @note This is different from scikit-learn's OneHotEncoder, which keeps all categories. * The output vectors are sparse. * + * When `handleInvalid` is configured to 'keep', an extra "category" indicating invalid values is + * added as last category. So when `dropLast` is true, invalid values are encoded as all-zeros + * vector. + * + * @note When encoding multi-column by using `inputCols` and `outputCols` params, input/output cols + * come in pairs, specified by the order in the arrays, and each pair is treated independently. + * * @see `StringIndexer` for converting categorical values into category indices - * @deprecated `OneHotEncoderEstimator` will be renamed `OneHotEncoder` and this `OneHotEncoder` - * will be removed in 3.0.0. */ -@Since("1.4.0") -@deprecated("`OneHotEncoderEstimator` will be renamed `OneHotEncoder` and this `OneHotEncoder`" + - " will be removed in 3.0.0.", "2.3.0") -class OneHotEncoder @Since("1.4.0") (@Since("1.4.0") override val uid: String) extends Transformer - with HasInputCol with HasOutputCol with DefaultParamsWritable { +@Since("2.3.0") +class OneHotEncoder @Since("2.3.0") (@Since("2.3.0") override val uid: String) --- End diff -- In this renaming case, I'm not sure we should use the `Since` from old `OneHotEncoder` (`1.4.0`) or `OneHotEncoderEstimator` (`2.3.0`). Now I use `OneHotEncoderEstimator`'s.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org