-
Notifications
You must be signed in to change notification settings - Fork 703
/
BigTextMatcherTestSpec.scala
210 lines (171 loc) · 7.43 KB
/
BigTextMatcherTestSpec.scala
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
/*
* 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.btm
import com.johnsnowlabs.nlp.AnnotatorType._
import com.johnsnowlabs.nlp._
import com.johnsnowlabs.nlp.annotators.Tokenizer
import com.johnsnowlabs.nlp.annotators.sbd.pragmatic.SentenceDetector
import com.johnsnowlabs.nlp.util.io.ReadAs
import com.johnsnowlabs.tags.FastTest
import org.apache.spark.ml.{Pipeline, PipelineModel}
import org.apache.spark.sql.{Dataset, Row}
import org.scalatest.flatspec.AnyFlatSpec
class BigTextMatcherTestSpec extends AnyFlatSpec with BigTextMatcherBehaviors {
"An BigTextMatcher" should s"be of type $CHUNK" taggedAs FastTest in {
val entityExtractor = new BigTextMatcherModel
assert(entityExtractor.outputAnnotatorType == CHUNK)
}
"A BigTextMatcher" should "extract entities with and without sentences" taggedAs FastTest in {
val dataset =
DataBuilder.basicDataBuild("Hello dolore magna aliqua. Lorem ipsum dolor. sit in laborum")
val result = AnnotatorBuilder.withFullBigTextMatcher(dataset)
val resultNoSentence = AnnotatorBuilder.withFullBigTextMatcher(dataset, sbd = false)
val resultNoSentenceNoCase =
AnnotatorBuilder.withFullBigTextMatcher(dataset, sbd = false, caseSensitive = false)
val extractedSentenced = Annotation.collect(result, "entity").flatten.toSeq
val extractedNoSentence = Annotation.collect(resultNoSentence, "entity").flatten.toSeq
val extractedNoSentenceNoCase =
Annotation.collect(resultNoSentenceNoCase, "entity").flatten.toSeq
val expectedSentenced = Seq(
Annotation(CHUNK, 6, 24, "dolore magna aliqua", Map("sentence" -> "0", "chunk" -> "0")),
Annotation(CHUNK, 53, 59, "laborum", Map("sentence" -> "2", "chunk" -> "1")))
val expectedNoSentence = Seq(
Annotation(CHUNK, 6, 24, "dolore magna aliqua", Map("sentence" -> "0", "chunk" -> "0")),
Annotation(CHUNK, 53, 59, "laborum", Map("sentence" -> "0", "chunk" -> "1")))
val expectedNoSentenceNoCase = Seq(
Annotation(CHUNK, 6, 24, "dolore magna aliqua", Map("sentence" -> "0", "chunk" -> "0")),
Annotation(CHUNK, 27, 48, "Lorem ipsum dolor. sit", Map("sentence" -> "0", "chunk" -> "1")),
Annotation(CHUNK, 53, 59, "laborum", Map("sentence" -> "0", "chunk" -> "2")))
assert(extractedSentenced == expectedSentenced)
assert(extractedNoSentence == expectedNoSentence)
assert(extractedNoSentenceNoCase == expectedNoSentenceNoCase)
}
"An Entity Extractor" should "search inside sentences" taggedAs FastTest in {
val dataset = DataBuilder.basicDataBuild("Hello dolore magna. Aliqua")
val result = AnnotatorBuilder.withFullBigTextMatcher(dataset, caseSensitive = false)
val extracted = Annotation.collect(result, "entity").flatten.toSeq
assert(extracted == Seq.empty[Annotation])
}
"A Recursive Pipeline BigTextMatcher" should "extract entities from dataset" taggedAs FastTest in {
val data = ContentProvider.parquetData.limit(1000)
val documentAssembler = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val sentenceDetector = new SentenceDetector()
.setInputCols(Array("document"))
.setOutputCol("sentence")
val tokenizer = new Tokenizer()
.setInputCols(Array("sentence"))
.setOutputCol("token")
val entityExtractor = new BigTextMatcher()
.setInputCols("sentence", "token")
.setStoragePath("src/test/resources/entity-extractor/test-phrases.txt", ReadAs.TEXT)
.setOutputCol("entity")
val finisher = new Finisher()
.setInputCols("entity")
.setOutputAsArray(false)
.setAnnotationSplitSymbol("@")
.setValueSplitSymbol("#")
val recursivePipeline = new Pipeline()
.setStages(Array(documentAssembler, sentenceDetector, tokenizer, entityExtractor, finisher))
val m = recursivePipeline.fit(data)
m.write.overwrite().save("./tmp_bigtm")
m.transform(data).show(1, truncate = false)
assert(recursivePipeline.fit(data).transform(data).filter("finished_entity == ''").count > 0)
}
"A big text matcher pipeline" should "work fine" taggedAs FastTest in {
val m = PipelineModel.load("./tmp_bigtm")
val dataset = DataBuilder.basicDataBuild("Hello dolore magna. Aliqua")
m.transform(dataset).show(1, truncate = false)
}
val latinBodyData: Dataset[Row] = DataBuilder.basicDataBuild(ContentProvider.latinBody)
"A full Normalizer pipeline with latin content" should behave like fullBigTextMatcher(
latinBodyData)
"A BigTextMatcher" should "also match substrings of entities" taggedAs FastTest in {
val data =
DataBuilder.basicDataBuild("patient has Lung Cancer", "patient Lung and Kidney Cancer")
val tokenizedData = AnnotatorBuilder.withTokenizer(data, sbd = false)
val entityExtractor = new BigTextMatcher()
.setInputCols("document", "token")
.setStoragePath("src/test/resources/entity-extractor/test-overlapping.txt", ReadAs.TEXT)
.setOutputCol("entity")
.setCaseSensitive(false)
val results = entityExtractor.fit(tokenizedData).transform(tokenizedData)
val collected = Annotation.collect(results, "entity").flatten
val expected = Seq(
Annotation(
CHUNK,
begin = 12,
end = 15,
result = "Lung",
Map("sentence" -> "0", "chunk" -> "0")),
Annotation(
CHUNK,
begin = 12,
end = 22,
result = "Lung Cancer",
Map("sentence" -> "0", "chunk" -> "1")),
Annotation(
CHUNK,
begin = 8,
end = 11,
result = "Lung",
Map("sentence" -> "0", "chunk" -> "0")),
Annotation(
CHUNK,
begin = 17,
end = 22,
result = "Kidney",
Map("sentence" -> "0", "chunk" -> "1")),
Annotation(
CHUNK,
begin = 17,
end = 29,
result = "Kidney Cancer",
Map("sentence" -> "0", "chunk" -> "2")))
assert(expected.length == collected.length)
expected.zip(collected).map { case (expAnno: Annotation, anno: Annotation) =>
assert(expAnno == anno)
}
// Test for merged chunks
entityExtractor.setMergeOverlapping(true)
val resultsMerged = entityExtractor.fit(tokenizedData).transform(tokenizedData)
val collectedMerged = Annotation.collect(resultsMerged, "entity").flatten
val expectedMerged = Seq(
Annotation(
CHUNK,
begin = 12,
end = 22,
result = "Lung Cancer",
Map("sentence" -> "0", "chunk" -> "0")),
Annotation(
CHUNK,
begin = 8,
end = 11,
result = "Lung",
Map("sentence" -> "0", "chunk" -> "0")),
Annotation(
CHUNK,
begin = 17,
end = 29,
result = "Kidney Cancer",
Map("sentence" -> "0", "chunk" -> "1")))
assert(expectedMerged.length == collectedMerged.length)
expectedMerged.zip(collectedMerged).map { case (expAnno: Annotation, anno: Annotation) =>
assert(expAnno == anno)
}
}
}