com.johnsnowlabs.nlp.annotators.cv.Gemma3ForMultiModal.scala Maven / Gradle / Ivy
/*
* Copyright 2017-2024 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.cv
import com.johnsnowlabs.ml.ai.util.Generation.GenerationConfig
import com.johnsnowlabs.ml.ai.Gemma3
import com.johnsnowlabs.ml.onnx.OnnxWrapper.DecoderWrappers
import com.johnsnowlabs.ml.util.LoadExternalModel.{
loadJsonStringAsset,
loadTextAsset,
modelSanityCheck,
notSupportedEngineError
}
import com.johnsnowlabs.nlp.annotators.cv.feature_extractor.Preprocessor
import com.johnsnowlabs.ml.util.Openvino
import com.johnsnowlabs.nlp.AnnotatorType.{DOCUMENT, IMAGE}
import com.johnsnowlabs.nlp._
import org.json4s.{DefaultFormats, JValue}
import org.json4s.jackson.JsonMethods.parse
import com.johnsnowlabs.ml.openvino.{OpenvinoWrapper, ReadOpenvinoModel, WriteOpenvinoModel}
import com.johnsnowlabs.ml.openvino.OpenvinoWrapper.Gemma3Wrappers
import com.johnsnowlabs.nlp.serialization.{MapFeature, StructFeature}
import org.apache.spark.broadcast.Broadcast
import org.apache.spark.ml.param.{IntArrayParam, IntParam}
import org.apache.spark.ml.util.Identifiable
import org.apache.spark.sql.SparkSession
/** Gemma3ForMultiModal can load Gemma3 Vision models for visual question answering. The model
* consists of a vision encoder, a text encoder, a text decoder and a model merger. The vision
* encoder will encode the input image, the text encoder will encode the input text, the model
* merger will merge the image and text embeddings, and the text decoder will output the answer.
*
* Gemma 3 is a family of lightweight, state-of-the-art open models from Google, built from the
* same research and technology used to create the Gemini models. Key features include:
* - Large 128K context window
* - Multilingual support in over 140 languages
* - Multimodal capabilities handling both text and image inputs
* - Optimized for deployment on limited resources (laptops, desktops, cloud)
*
* Pretrained models can be loaded with `pretrained` of the companion object:
* {{{
* val visualQA = Gemma3ForMultiModal.pretrained()
* .setInputCols("image_assembler")
* .setOutputCol("answer")
* }}}
* The default model is `"gemma3_4b_it_int4"`, if no name is provided.
*
* For available pretrained models please see the
* [[https://sparknlp.org/models?task=Question+Answering Models Hub]].
*
* ==Example==
* {{{
* import spark.implicits._
* import com.johnsnowlabs.nlp.base._
* import com.johnsnowlabs.nlp.annotator._
* import org.apache.spark.ml.Pipeline
*
* val imageDF = spark.read
* .format("image")
* .option("dropInvalid", value = true)
* .load(imageFolder)
*
* val testDF = imageDF.withColumn("text", lit("user\nYou are a helpful assistant.\n\nDescribe this image in detail.\nmodel\n"))
*
* val imageAssembler = new ImageAssembler()
* .setInputCol("image")
* .setOutputCol("image_assembler")
*
* val visualQA = Gemma3ForMultiModal.pretrained()
* .setInputCols("image_assembler")
* .setOutputCol("answer")
*
* val pipeline = new Pipeline().setStages(Array(
* imageAssembler,
* visualQA
* ))
*
* val result = pipeline.fit(testDF).transform(testDF)
*
* result.select("image_assembler.origin", "answer.result").show(truncate = false)
* }}}
*
* @param uid
* required uid for storing annotator to disk
*/
class Gemma3ForMultiModal(override val uid: String)
extends AnnotatorModel[Gemma3ForMultiModal]
with HasBatchedAnnotateImage[Gemma3ForMultiModal]
with HasImageFeatureProperties
with WriteOpenvinoModel
with HasGeneratorProperties
with HasEngine {
/** Annotator reference id. Used to identify elements in metadata or to refer to this annotator
* type
*/
def this() = this(Identifiable.randomUID("Gemma3ForMultiModal"))
/** Annotator reference id. Used to identify elements in metadata or to refer to this annotator
* type
*/
override val inputAnnotatorTypes: Array[AnnotatorType] = Array(IMAGE)
override val outputAnnotatorType: AnnotatorType = DOCUMENT
/** @group setParam */
def setRandomSeed(value: Int): Gemma3ForMultiModal.this.type = {
if (randomSeed.isEmpty) {
this.randomSeed = Some(value)
}
this
}
/** A list of token ids which are ignored in the decoder's output (Default: `Array()`)
*
* @group param
*/
var ignoreTokenIds = new IntArrayParam(
this,
"ignoreTokenIds",
"A list of token ids which are ignored in the decoder's output")
/** @group setParam */
def setIgnoreTokenIds(tokenIds: Array[Int]): Gemma3ForMultiModal.this.type = {
set(ignoreTokenIds, tokenIds)
}
/** @group getParam */
def getIgnoreTokenIds: Array[Int] = $(ignoreTokenIds)
/** Vocabulary used to encode the words to ids with bpeTokenizer.encode
*
* @group param
*/
val vocabulary: MapFeature[String, Int] = new MapFeature(this, "vocabulary").setProtected()
/** @group setParam */
def setVocabulary(value: Map[String, Int]): this.type = set(vocabulary, value)
/** Holding merges.txt coming from RoBERTa model
*
* @group param
*/
val merges: MapFeature[(String, String), Int] = new MapFeature(this, "merges").setProtected()
/** @group setParam */
def setMerges(value: Map[(String, String), Int]): this.type = set(merges, value)
/** Additional tokens to be added to the vocabulary
*
* @group param
*/
val addedTokens: MapFeature[String, Int] = new MapFeature(this, "addedTokens").setProtected()
/** @group setParam */
def setAddedTokens(value: Map[String, Int]): this.type = set(addedTokens, value)
/** Stop tokens to terminate the generation
*
* @group param
*/
override val stopTokenIds =
new IntArrayParam(this, "stopTokenIds", "Stop tokens to terminate the generation")
/** @group setParam */
override def setStopTokenIds(value: Array[Int]): this.type = {
set(stopTokenIds, value)
}
/** @group getParam */
override def getStopTokenIds: Array[Int] = $(stopTokenIds)
private var _model: Option[Broadcast[Gemma3]] = None
val generationConfig: StructFeature[GenerationConfig] =
new StructFeature(this, "generationConfig").setProtected()
def setGenerationConfig(value: GenerationConfig): this.type =
set(generationConfig, value)
def getGenerationConfig: GenerationConfig = $$(generationConfig)
val imageToken =
new IntParam(this, "imageToken", "Token id for image embeddings")
/** @group setParam */
def setImageToken(value: Int): this.type = set(imageToken, value)
/** @group getParam */
def getImageToken: Int = $(imageToken)
val imageTokenLength =
new IntParam(this, "imageTokenLength", "Token length for image embeddings")
/** @group setParam */
def setImageTokenLength(value: Int): this.type = set(imageTokenLength, value)
/** @group getParam */
def getImageTokenLength: Int = $(imageTokenLength)
/** @group setParam */
def setModelIfNotSet(
spark: SparkSession,
preprocessor: Preprocessor,
onnxWrappers: Option[DecoderWrappers],
openvinoWrapper: Option[Gemma3Wrappers]): this.type = {
if (_model.isEmpty) {
_model = Some(
spark.sparkContext.broadcast(
new Gemma3(
onnxWrappers,
openvinoWrapper,
$$(merges),
$$(vocabulary),
$$(addedTokens),
preprocessor,
generationConfig = getGenerationConfig,
imageToken = getImageToken,
imageTokenLength = getImageTokenLength)))
}
this
}
/** @group getParam */
def getModelIfNotSet: Gemma3 = _model.get.value
setDefault(
minOutputLength -> 0,
maxOutputLength -> 20,
doSample -> false,
temperature -> 0.6,
topK -> -1,
topP -> 0.9,
repetitionPenalty -> 1.0,
noRepeatNgramSize -> 3,
ignoreTokenIds -> Array(),
batchSize -> 1,
beamSize -> 1,
maxInputLength -> 4096,
stopTokenIds -> Array(2),
imageToken -> 257152,
imageTokenLength -> 256)
/** takes a document and annotations and produces new annotations of this annotator's annotation
* type
*
* @param batchedAnnotations
* Annotations in batches that correspond to inputAnnotationCols generated by previous
* annotators if any
* @return
* any number of annotations processed for every batch of input annotations. Not necessary
* one to one relationship
*/
override def batchAnnotate(
batchedAnnotations: Seq[Array[AnnotationImage]]): Seq[Seq[Annotation]] = {
batchedAnnotations.map { cleanAnnotationImages =>
val validImages = cleanAnnotationImages.filter(_.result.nonEmpty)
val questionAnnotations = extractInputAnnotation(validImages)
getModelIfNotSet.predict(
questionAnnotations,
validImages.toSeq,
batchSize = $(batchSize),
minOutputLength = $(minOutputLength),
maxOutputLength = $(maxOutputLength),
doSample = $(doSample),
temperature = $(temperature),
topK = $(topK),
topP = $(topP),
repetitionPenalty = $(repetitionPenalty),
noRepeatNgramSize = $(noRepeatNgramSize),
randomSeed = this.randomSeed,
ignoreTokenIds = $(ignoreTokenIds),
beamSize = $(beamSize),
maxInputLength = $(maxInputLength))
}
}
private def extractInputAnnotation(
annotationImages: Array[AnnotationImage]): Seq[Annotation] = {
val questions = annotationImages.map(annotationImage => {
val imageText =
if (annotationImage.text.nonEmpty) annotationImage.text
else
"user\nYou are a helpful assistant.\n\nDescribe this image in detail.\nmodel\n" // default question
Annotation(imageText)
})
questions
}
override def onWrite(path: String, spark: SparkSession): Unit = {
super.onWrite(path, spark)
getEngine match {
case Openvino.name =>
val wrappers = getModelIfNotSet.openvinoWrapper
writeOpenvinoModels(
path,
spark,
Seq((wrappers.get.languageModel, "openvino_language_model.xml")),
Gemma3ForMultiModal.suffix)
writeOpenvinoModels(
path,
spark,
Seq((wrappers.get.imageEncoder, "openvino_vision_embeddings_model.xml")),
Gemma3ForMultiModal.suffix)
writeOpenvinoModels(
path,
spark,
Seq((wrappers.get.textEmbeddings, "openvino_text_embeddings_model.xml")),
Gemma3ForMultiModal.suffix)
writeOpenvinoModels(
path,
spark,
Seq((wrappers.get.modelMerger, "openvino_merger_model.xml")),
Gemma3ForMultiModal.suffix)
case _ =>
throw new Exception(notSupportedEngineError)
}
}
}
trait ReadablePretrainedGemma3ForMultiModal
extends ParamsAndFeaturesReadable[Gemma3ForMultiModal]
with HasPretrained[Gemma3ForMultiModal] {
override val defaultModelName: Some[String] = Some("gemma3_4b_it_int4")
/** Java compliant-overrides */
override def pretrained(): Gemma3ForMultiModal = super.pretrained()
override def pretrained(name: String): Gemma3ForMultiModal =
super.pretrained(name)
override def pretrained(name: String, lang: String): Gemma3ForMultiModal =
super.pretrained(name, lang)
override def pretrained(name: String, lang: String, remoteLoc: String): Gemma3ForMultiModal =
super.pretrained(name, lang, remoteLoc)
}
trait ReadGemma3ForMultiModalDLModel extends ReadOpenvinoModel {
this: ParamsAndFeaturesReadable[Gemma3ForMultiModal] =>
val suffix: String = "_Gemma3"
override val openvinoFile: String = "Gemma3_openvino"
def readModel(instance: Gemma3ForMultiModal, path: String, spark: SparkSession): Unit = {
instance.getEngine match {
case Openvino.name =>
val languageModelWrappers =
readOpenvinoModels(path, spark, Seq("openvino_language_model.xml"), suffix)
val imageEncoderWrappers =
readOpenvinoModels(path, spark, Seq("openvino_vision_embeddings_model.xml"), suffix)
val textEmbeddingsWrappers =
readOpenvinoModels(path, spark, Seq("openvino_text_embeddings_model.xml"), suffix)
val modelMergerWrappers =
readOpenvinoModels(path, spark, Seq("openvino_merger_model.xml"), suffix)
val ovWrapper = Gemma3Wrappers(
languageModel = languageModelWrappers("openvino_language_model.xml"),
imageEncoder = imageEncoderWrappers("openvino_vision_embeddings_model.xml"),
textEmbeddings = textEmbeddingsWrappers("openvino_text_embeddings_model.xml"),
modelMerger = modelMergerWrappers("openvino_merger_model.xml"))
val preprocessor = Preprocessor(
do_normalize = true,
do_resize = true,
"Gemma3FeatureExtractor",
instance.getImageMean,
instance.getImageStd,
instance.getResample,
instance.getSize)
instance.setModelIfNotSet(spark, preprocessor, None, Some(ovWrapper))
case _ => {
throw new Exception(notSupportedEngineError)
}
}
}
addReader(readModel)
def loadSavedModel(
modelPath: String,
spark: SparkSession,
useOpenvino: Boolean = false): Gemma3ForMultiModal = {
implicit val formats: DefaultFormats.type = DefaultFormats // for json4s
val (localModelPath, detectedEngine) =
modelSanityCheck(
modelPath,
isDecoder = false,
custom = Some(
List(
"openvino_language_model",
"openvino_vision_embeddings_model",
"openvino_text_embeddings_model",
"openvino_merger_model")))
val modelConfig: JValue =
parse(loadJsonStringAsset(localModelPath, "config.json"))
val generationConfigJson: JValue = parse(
loadJsonStringAsset(localModelPath, "generation_config.json"))
val preprocessorConfigJsonContent =
loadJsonStringAsset(localModelPath, "preprocessor_config.json")
val preprocessorConfig = Preprocessor.loadPreprocessorConfig(preprocessorConfigJsonContent)
val beginSuppressTokens: Array[Int] =
(modelConfig \ "begin_suppress_tokens").extract[Array[Int]]
val suppressTokenIds: Array[Int] =
(modelConfig \ "suppress_tokens").extract[Array[Int]]
val forcedDecoderIds: Array[(Int, Int)] =
(modelConfig \ "forced_decoder_ids").extract[Array[Array[Int]]].map {
case idxWithTokenId: Array[Int] if idxWithTokenId.length == 2 =>
(idxWithTokenId(0), idxWithTokenId(1))
case _ =>
throw new Exception(
"Could not extract forced_decoder_ids. Should be a list of tuples with 2 entries.")
}
def arrayOrNone[T](array: Array[T]): Option[Array[T]] =
if (array.nonEmpty) Some(array) else None
val bosTokenId = (generationConfigJson \ "bos_token_id").extract[Int]
val eosTokenIdArray = (generationConfigJson \ "eos_token_id").extract[Array[Int]]
val eosTokenId = eosTokenIdArray.head
// val eosTokenId = (generationConfigJson \ "eos_token_id").extract[Int]
val padTokenId = (generationConfigJson \ "pad_token_id").extract[Int]
val vocabSize = (modelConfig \ "text_config" \ "vocab_size").extract[Int]
val imageToken = (modelConfig \ "image_token_index").extract[Int]
val imageTokenLength = (modelConfig \ "mm_tokens_per_image").extract[Int]
// Check if tokenizer.json exists
val tokenizerPath = s"$localModelPath/assets/tokenizer.json"
val tokenizerExists = new java.io.File(tokenizerPath).exists()
val (vocabs, addedTokens, bytePairs) = if (tokenizerExists) {
val tokenizerConfig: JValue = parse(loadJsonStringAsset(localModelPath, "tokenizer.json"))
var vocabs: Map[String, Int] =
(tokenizerConfig \ "model" \ "vocab").extract[Map[String, Int]]
val bytePairs = (tokenizerConfig \ "model" \ "merges")
.extract[List[Array[String]]]
.filter(w => w.length == 2)
.map { case Array(c1, c2) => (c1, c2) }
.zipWithIndex
.toMap
val addedTokens = (tokenizerConfig \ "added_tokens")
.extract[List[Map[String, Any]]]
.map { token =>
val id = token("id").asInstanceOf[BigInt].intValue()
val content = token("content").asInstanceOf[String]
(content, id)
}
.toMap
addedTokens.foreach { case (content, id) =>
vocabs += (content -> id)
}
(vocabs, addedTokens, bytePairs)
} else {
val vocabs = loadTextAsset(localModelPath, "vocab.txt").zipWithIndex.toMap
val addedTokens = loadTextAsset(localModelPath, "added_tokens.txt").zipWithIndex.toMap
val bytePairs = loadTextAsset(localModelPath, "merges.txt")
.map(_.split(" "))
.filter(w => w.length == 2)
.map { case Array(c1, c2) => (c1, c2) }
.zipWithIndex
.toMap
(vocabs, addedTokens, bytePairs)
}
val annotatorModel = new Gemma3ForMultiModal()
.setGenerationConfig(
GenerationConfig(
bosTokenId,
padTokenId,
eosTokenId,
vocabSize,
arrayOrNone(beginSuppressTokens),
arrayOrNone(suppressTokenIds),
arrayOrNone(forcedDecoderIds)))
.setVocabulary(vocabs)
.setMerges(bytePairs)
.setAddedTokens(addedTokens)
.setImageToken(imageToken)
.setImageTokenLength(imageTokenLength)
.setSize(preprocessorConfig.size)
.setImageMean(preprocessorConfig.image_mean)
.setImageStd(preprocessorConfig.image_std)
.setResample(preprocessorConfig.resample)
.setStopTokenIds(eosTokenIdArray)
val modelEngine =
if (useOpenvino)
Openvino.name
else
detectedEngine
annotatorModel.set(annotatorModel.engine, modelEngine)
detectedEngine match {
case Openvino.name =>
val imageEncoderWrapper =
OpenvinoWrapper.read(
spark,
localModelPath,
zipped = false,
useBundle = true,
detectedEngine = detectedEngine,
modelName = "openvino_vision_embeddings_model")
val textEmbeddingsWrapper =
OpenvinoWrapper.read(
spark,
localModelPath,
zipped = false,
useBundle = true,
detectedEngine = detectedEngine,
modelName = "openvino_text_embeddings_model")
val modelMergerWrapper =
OpenvinoWrapper.read(
spark,
localModelPath,
zipped = false,
useBundle = true,
detectedEngine = detectedEngine,
modelName = "openvino_merger_model")
val languageModelWrapper =
OpenvinoWrapper.read(
spark,
localModelPath,
zipped = false,
useBundle = true,
detectedEngine = detectedEngine,
modelName = "openvino_language_model")
val openvinoWrapper = Gemma3Wrappers(
languageModel = languageModelWrapper,
imageEncoder = imageEncoderWrapper,
textEmbeddings = textEmbeddingsWrapper,
modelMerger = modelMergerWrapper)
annotatorModel.setModelIfNotSet(spark, preprocessorConfig, None, Some(openvinoWrapper))
case _ =>
throw new Exception(notSupportedEngineError)
}
annotatorModel
}
}
object Gemma3ForMultiModal
extends ReadablePretrainedGemma3ForMultiModal
with ReadGemma3ForMultiModalDLModel
© 2015 - 2025 Weber Informatics LLC | Privacy Policy