SPARKNLP-835: ProtectedParam and ProtectedFeature (#13797)

* SPARKNLP-835: Finalize protected Features

* SPARKNLP-835: Remove redundant checks for protected Features

* SPARKNLP-835: Introduce ProtectedParam

* SPARKNLP-835: Resolve encoding/decoding issue for HasProtectedParams

* SPARKNLP-835: Make caseSensitive settable

* SPARKNLP-835: Make maxSentenceLength, batchSize settable

* SPARKNLP-835: Enable protected Params for Annotators

Author: DevinTDHa

src/main/scala/com/johnsnowlabs/nlp/HasProtectedParams.scala

@@ -0,0 +1,62 @@
+package com.johnsnowlabs.nlp
+
+import org.apache.spark.ml.param.{Param, Params}
+
+/** Enables a class to protect a parameter, which means that it can only be set once.
+  *
+  * This trait will enable a implicit conversion from Param to ProtectedParam. In addition, the
+  * new set for ProtectedParam will then check, whether or not the value was already set. If so,
+  * then a warning will be output and the value will not be set again.
+  */
+trait HasProtectedParams {
+  this: Params =>
+  implicit class ProtectedParam[T](baseParam: Param[T])
+      extends Param[T](baseParam.parent, baseParam.name, baseParam.doc, baseParam.isValid) {
+
+    var isProtected = false
+
+    /** Sets this parameter to be protected, which means that it can only be set once.
+      *
+      * Default values do not count as a set value and can be overridden.
+      *
+      * @return
+      *   This object
+      */
+    def setProtected(): this.type = {
+      isProtected = true
+      this
+    }
+
+    def toParam: Param[T] = this.asInstanceOf[Param[T]]
+
+    // Overrides needed for individual Param implementation
+    override def jsonEncode(value: T): String = baseParam.jsonEncode(value)
+    override def jsonDecode(json: String): T = baseParam.jsonDecode(json)
+  }
+
+  /** Sets the value for a protected Param.
+    *
+    * If the parameter was already set, it will not be set again. Default values do not count as a
+    * set value and can be overridden.
+    *
+    * @param param
+    *   Protected parameter to set
+    * @param value
+    *   Value for the parameter
+    * @tparam T
+    *   Type of the parameter
+    * @return
+    *   This object
+    */
+  def set[T](param: ProtectedParam[T], value: T): this.type = {
+    if (param.isProtected && get(param).isDefined)
+      println(
+        s"Warning: The parameter ${param.name} is protected and can only be set once." +
+          " For a pretrained model, this was done during the initialization process." +
+          " If you are trying to train your own model, please check the documentation." +
+          " If this is intentional, set the parameter directly with set(annotator.param, value).")
+    else
+      set(param.toParam, value)
+    this
+  }
+}

src/main/scala/com/johnsnowlabs/nlp/annotators/audio/Wav2Vec2ForCTC.scala

@@ -189,8 +189,7 @@ class Wav2Vec2ForCTC(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/AlbertForQuestionAnswering.scala

@@ -184,8 +184,7 @@ class AlbertForQuestionAnswering(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/AlbertForSequenceClassification.scala

@@ -227,8 +227,7 @@ class AlbertForSequenceClassification(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 
@@ -265,9 +264,7 @@ class AlbertForSequenceClassification(override val uid: String)
     * @group setParam
     */
   override def setCaseSensitive(value: Boolean): this.type = {
-    if (get(caseSensitive).isEmpty)
-      set(this.caseSensitive, value)
-    this
+    set(this.caseSensitive, value)
   }
 
   setDefault(

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/AlbertForTokenClassification.scala

@@ -205,8 +205,7 @@ class AlbertForTokenClassification(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 
@@ -242,9 +241,7 @@ class AlbertForTokenClassification(override val uid: String)
     * @group setParam
     */
   override def setCaseSensitive(value: Boolean): this.type = {
-    if (get(caseSensitive).isEmpty)
-      set(this.caseSensitive, value)
-    this
+    set(this.caseSensitive, value)
   }
 
   setDefault(batchSize -> 8, maxSentenceLength -> 128, caseSensitive -> false)

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/BertForQuestionAnswering.scala

@@ -198,8 +198,7 @@ class BertForQuestionAnswering(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/BertForSequenceClassification.scala

@@ -243,8 +243,7 @@ class BertForSequenceClassification(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 
@@ -282,9 +281,7 @@ class BertForSequenceClassification(override val uid: String)
     * @group setParam
     */
   override def setCaseSensitive(value: Boolean): this.type = {
-    if (get(caseSensitive).isEmpty)
-      set(this.caseSensitive, value)
-    this
+    set(this.caseSensitive, value)
   }
 
   setDefault(

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/BertForTokenClassification.scala

@@ -217,8 +217,7 @@ class BertForTokenClassification(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 
@@ -255,9 +254,7 @@ class BertForTokenClassification(override val uid: String)
     * @group setParam
     */
   override def setCaseSensitive(value: Boolean): this.type = {
-    if (get(caseSensitive).isEmpty)
-      set(this.caseSensitive, value)
-    this
+    set(this.caseSensitive, value)
   }
 
   setDefault(batchSize -> 8, maxSentenceLength -> 128, caseSensitive -> true)

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/BertForZeroShotClassification.scala

@@ -167,8 +167,7 @@ class BertForZeroShotClassification(override val uid: String)
 
   /** @group setParam */
   def setVocabulary(value: Map[String, Int]): this.type = {
-    if (get(vocabulary).isEmpty)
-      set(vocabulary, value)
+    set(vocabulary, value)
     this
   }
 
@@ -256,8 +255,7 @@ class BertForZeroShotClassification(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }
 
@@ -295,9 +293,7 @@ class BertForZeroShotClassification(override val uid: String)
     * @group setParam
     */
   override def setCaseSensitive(value: Boolean): this.type = {
-    if (get(caseSensitive).isEmpty)
-      set(this.caseSensitive, value)
-    this
+    set(this.caseSensitive, value)
   }
 
   setDefault(

src/main/scala/com/johnsnowlabs/nlp/annotators/classifier/dl/CamemBertForQuestionAnswering.scala

@@ -184,8 +184,7 @@ class CamemBertForQuestionAnswering(override val uid: String)
 
   /** @group setParam */
   def setSignatures(value: Map[String, String]): this.type = {
-    if (get(signatures).isEmpty)
-      set(signatures, value)
+    set(signatures, value)
     this
   }