Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/9513#discussion_r44311160
--- Diff: mllib/src/main/scala/org/apache/spark/ml/clustering/LDA.scala ---
@@ -0,0 +1,740 @@
+/*
+ * 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.clustering
+
+import org.apache.spark.Logging
+import org.apache.spark.annotation.{Experimental, Since}
+import org.apache.spark.ml.util.{SchemaUtils, Identifiable}
+import org.apache.spark.ml.{Estimator, Model}
+import org.apache.spark.ml.param.shared.{HasCheckpointInterval,
HasFeaturesCol, HasSeed, HasMaxIter}
+import org.apache.spark.ml.param._
+import org.apache.spark.mllib.clustering.{DistributedLDAModel =>
OldDistributedLDAModel,
+ EMLDAOptimizer => OldEMLDAOptimizer, LDA => OldLDA, LDAModel =>
OldLDAModel,
+ LDAOptimizer => OldLDAOptimizer, LocalLDAModel => OldLocalLDAModel,
+ OnlineLDAOptimizer => OldOnlineLDAOptimizer}
+import org.apache.spark.mllib.linalg.{VectorUDT, Vectors, Matrix, Vector}
+import org.apache.spark.rdd.RDD
+import org.apache.spark.sql.{SQLContext, DataFrame, Row}
+import org.apache.spark.sql.functions.{col, monotonicallyIncreasingId, udf}
+import org.apache.spark.sql.types.StructType
+
+
+private[clustering] trait LDAParams extends Params with HasFeaturesCol
with HasMaxIter
+ with HasSeed with HasCheckpointInterval {
+
+ /**
+ * Param for the number of topics (clusters). Must be > 1. Default: 10.
+ * @group param
+ */
+ @Since("1.6.0")
+ final val k = new IntParam(this, "k", "number of clusters to create",
ParamValidators.gt(1))
+
+ /** @group getParam */
+ @Since("1.6.0")
+ def getK: Int = $(k)
+
+ /**
+ * Concentration parameter (commonly named "alpha") for the prior placed
on documents'
+ * distributions over topics ("theta").
+ *
+ * This is the parameter to a Dirichlet distribution, where larger
values mean more smoothing
+ * (more regularization).
+ *
+ * If set to a singleton vector [-1], then docConcentration is set
automatically. If set to
+ * singleton vector [alpha] where alpha != -1, then alpha is replicated
to a vector of
+ * length k in fitting. Otherwise, the [[docConcentration]] vector must
be length k.
+ * (default = [-1] = automatic)
+ *
+ * Optimizer-specific parameter settings:
+ * - EM
+ * - Currently only supports symmetric distributions, so all values
in the vector should be
+ * the same.
+ * - Values should be > 1.0
+ * - default = uniformly (50 / k) + 1, where 50/k is common in LDA
libraries and +1 follows
+ * from Asuncion et al. (2009), who recommend a +1 adjustment for
EM.
+ * - Online
+ * - Values should be >= 0
+ * - default = uniformly (1.0 / k), following the implementation from
+ * [[https://github.com/Blei-Lab/onlineldavb]].
+ * @group param
+ */
+ @Since("1.6.0")
+ final val docConcentration = new DoubleArrayParam(this,
"docConcentration",
+ "Concentration parameter (commonly named \"alpha\") for the prior
placed on documents'" +
+ " distributions over topics (\"theta\").", validDocConcentration)
+
+ /** Check that the docConcentration is valid, independently of other
Params */
+ private def validDocConcentration(alpha: Array[Double]): Boolean = {
+ if (alpha.length == 1) {
+ alpha(0) == -1 || alpha(0) >= 1.0
+ } else if (alpha.length > 1) {
+ alpha.forall(_ >= 1.0)
+ } else {
+ false
+ }
+ }
+
+ /** @group getParam */
+ @Since("1.6.0")
+ def getDocConcentration: Array[Double] = $(docConcentration)
+
+ /**
+ * Alias for [[getDocConcentration]]
+ * @group getParam
+ */
+ @Since("1.6.0")
+ def getAlpha: Array[Double] = getDocConcentration
+
+ /**
+ * Concentration parameter (commonly named "beta" or "eta") for the
prior placed on topics'
+ * distributions over terms.
+ *
+ * This is the parameter to a symmetric Dirichlet distribution.
+ *
+ * Note: The topics' distributions over terms are called "beta" in the
original LDA paper
+ * by Blei et al., but are called "phi" in many later papers such as
Asuncion et al., 2009.
+ *
+ * If set to -1, then topicConcentration is set automatically.
+ * (default = -1 = automatic)
+ *
+ * Optimizer-specific parameter settings:
+ * - EM
+ * - Value should be > 1.0
+ * - default = 0.1 + 1, where 0.1 gives a small amount of smoothing
and +1 follows
+ * Asuncion et al. (2009), who recommend a +1 adjustment for EM.
+ * - Online
+ * - Value should be >= 0
+ * - default = (1.0 / k), following the implementation from
+ * [[https://github.com/Blei-Lab/onlineldavb]].
+ * @group param
+ */
+ @Since("1.6.0")
+ final val topicConcentration = new DoubleParam(this,
"topicConcentration",
+ "Concentration parameter (commonly named \"beta\" or \"eta\") for the
prior placed on topic'" +
+ " distributions over terms.", (beta: Double) => beta == -1 || beta
>= 0.0)
+
+ /** @group getParam */
+ @Since("1.6.0")
+ def getTopicConcentration: Double = $(topicConcentration)
+
+ /**
+ * Alias for [[getTopicConcentration]]
+ * @group getParam
+ */
+ @Since("1.6.0")
+ def getBeta: Double = getTopicConcentration
+
+ /**
+ * Optimizer or inference algorithm used to estimate the LDA model,
specified as a
+ * [[LDAOptimizer]] type.
+ * Currently supported:
+ * - Online Variational Bayes: [[OnlineLDAOptimizer]] (default)
--- End diff --
I'm OK with debating this, but I think online may be better for most use
cases. What do you think? I think it's Ok to change behavior between APIs, if
it's for a good enough reason.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
