4413868535

Build Type

push

github

Committed by GitHub

Commit Message

SPARKNLP-746: Handle empty validation sets (#13615)

Run Details

8597 of 12936 relevant lines covered (66.46%)

0.66 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/src/main/scala/com/johnsnowlabs/nlp/annotators/common/WordpieceTokenized.scala

/*
 * Copyright 2017-2022 John Snow Labs
 *
 * Licensed 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 com.johnsnowlabs.nlp.annotators.common

import com.johnsnowlabs.nlp.{Annotation, AnnotatorType}

import scala.collection.Map

object WordpieceTokenized extends Annotated[WordpieceTokenizedSentence] {

  override def annotatorType: String = AnnotatorType.WORDPIECE

  override def unpack(annotations: Seq[Annotation]): Seq[WordpieceTokenizedSentence] = {
    val tokens = annotations
      .filter(_.annotatorType == annotatorType)
      .toArray

    SentenceSplit
      .unpack(annotations)
      .map(sentence => {
        tokens
          .filter(token => token.begin >= sentence.start & token.end <= sentence.end)
          .map(token =>
            TokenPiece(
              wordpiece = token.result,
              token = token.metadata("token"),
              pieceId = token.metadata("pieceId").toInt,
              isWordStart = token.metadata("isWordStart").toBoolean,
              begin = token.begin,
              end = token.end))
      })
      .filter(_.nonEmpty)
      .map(tokens => WordpieceTokenizedSentence(tokens))

  }

  override def pack(sentences: Seq[WordpieceTokenizedSentence]): Seq[Annotation] = {
    var sentenceIndex = 0

    sentences.flatMap { sentence =>
      sentenceIndex += 1
      sentence.tokens.map { token =>
        Annotation(
          annotatorType,
          token.begin,
          token.end,
          token.wordpiece,
          Map(
            "sentence" -> sentenceIndex.toString,
            "isWordStart" -> token.isWordStart.toString,
            "pieceId" -> token.pieceId.toString,
            "token" -> token.token))
      }
    }
  }
}

case class WordpieceTokenizedSentence(tokens: Array[TokenPiece])
case class TokenPiece(
    wordpiece: String,
    token: String,
    pieceId: Int,
    isWordStart: Boolean,
    begin: Int,
    end: Int)

1	/*
2	* Copyright 2017-2022 John Snow Labs
3	*
4	* Licensed under the Apache License, Version 2.0 (the "License");
5	* you may not use this file except in compliance with the License.
6	* You may obtain a copy of the License at
7	*
8	* http://www.apache.org/licenses/LICENSE-2.0
9	*
10	* Unless required by applicable law or agreed to in writing, software
11	* distributed under the License is distributed on an "AS IS" BASIS,
12	* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13	* See the License for the specific language governing permissions and
14	* limitations under the License.
15	*/
16
17	package com.johnsnowlabs.nlp.annotators.common
18
19	import com.johnsnowlabs.nlp.{Annotation, AnnotatorType}
20
21	import scala.collection.Map
22
23	object WordpieceTokenized extends Annotated[WordpieceTokenizedSentence] {
24
25	override def annotatorType: String = AnnotatorType.WORDPIECE	×
26
27	override def unpack(annotations: Seq[Annotation]): Seq[WordpieceTokenizedSentence] = {
28	val tokens = annotations
29	.filter(_.annotatorType == annotatorType)	×
30	.toArray	×
31
32	SentenceSplit
33	.unpack(annotations)
34	.map(sentence => {	×
35	tokens
36	.filter(token => token.begin >= sentence.start & token.end <= sentence.end)	×
37	.map(token =>	×
38	TokenPiece(	×
39	wordpiece = token.result,	×
40	token = token.metadata("token"),	×
41	pieceId = token.metadata("pieceId").toInt,	×
42	isWordStart = token.metadata("isWordStart").toBoolean,	×
43	begin = token.begin,	×
44	end = token.end))	×
45	})
46	.filter(_.nonEmpty)	×
47	.map(tokens => WordpieceTokenizedSentence(tokens))	×
48
49	}
50
51	override def pack(sentences: Seq[WordpieceTokenizedSentence]): Seq[Annotation] = {
52	var sentenceIndex = 0	×
53
54	sentences.flatMap { sentence =>	×
55	sentenceIndex += 1	×
56	sentence.tokens.map { token =>	×
57	Annotation(	×
58	annotatorType,	×
59	token.begin,	×
60	token.end,	×
61	token.wordpiece,	×
62	Map(	×
63	"sentence" -> sentenceIndex.toString,	×
64	"isWordStart" -> token.isWordStart.toString,	×
65	"pieceId" -> token.pieceId.toString,	×
66	"token" -> token.token))	×
67	}
68	}
69	}
70	}
71
72	case class WordpieceTokenizedSentence(tokens: Array[TokenPiece])
73	case class TokenPiece(
74	wordpiece: String,
75	token: String,
76	pieceId: Int,
77	isWordStart: Boolean,
78	begin: Int,
79	end: Int)

JohnSnowLabs / spark-nlp / 4413868535

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous