com.johnsnowlabs.ml.ai.DistilBert.scala Maven / Gradle / Ivy
The newest version!
/*
* 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.ml.ai
import ai.onnxruntime.OnnxTensor
import com.johnsnowlabs.ml.ai.util.PrepareEmbeddings
import com.johnsnowlabs.ml.onnx.{OnnxSession, OnnxWrapper}
import com.johnsnowlabs.ml.tensorflow.sign.{ModelSignatureConstants, ModelSignatureManager}
import com.johnsnowlabs.ml.tensorflow.{TensorResources, TensorflowWrapper}
import com.johnsnowlabs.ml.util.{ModelArch, ONNX, TensorFlow}
import com.johnsnowlabs.nlp.annotators.common._
import com.johnsnowlabs.nlp.{Annotation, AnnotatorType}
import org.slf4j.{Logger, LoggerFactory}
import scala.collection.JavaConverters._
/** The DistilBERT model was proposed in the paper '''DistilBERT, a distilled version of BERT:
* smaller, faster, cheaper and lighter''' [[https://arxiv.org/abs/1910.01108]]. DistilBERT is a
* small, fast, cheap and light Transformer model trained by distilling BERT base. It has 40%
* less parameters than `bert-base-uncased`, runs 60% faster while preserving over 95% of BERT's
* performances as measured on the GLUE language understanding benchmark.
*
* The abstract from the paper is the following:
*
* As Transfer Learning from large-scale pre-trained models becomes more prevalent in Natural
* Language Processing (NLP), operating these large models in on-the-edge and/or under
* constrained computational training or inference budgets remains challenging. In this work, we
* propose a method to pre-train a smaller general-purpose language representation model, called
* DistilBERT, which can then be fine-tuned with good performances on a wide range of tasks like
* its larger counterparts. While most prior work investigated the use of distillation for
* building task-specific models, we leverage knowledge distillation during the pretraining phase
* and show that it is possible to reduce the size of a BERT model by 40%, while retaining 97% of
* its language understanding capabilities and being 60% faster. To leverage the inductive biases
* learned by larger models during pretraining, we introduce a triple loss combining language
* modeling, distillation and cosine-distance losses. Our smaller, faster and lighter model is
* cheaper to pre-train and we demonstrate its capabilities for on-device computations in a
* proof-of-concept experiment and a comparative on-device study.
*
* Tips:
*
* - DistilBERT doesn't have :obj:`token_type_ids`, you don't need to indicate which token
* belongs to which segment. Just separate your segments with the separation token
* :obj:`tokenizer.sep_token` (or :obj:`[SEP]`).
*
* - DistilBERT doesn't have options to select the input positions (:obj:`position_ids` input).
* This could be added if necessary though, just let us know if you need this option.
*
* @param tensorflowWrapper
* Bert Model wrapper with TensorFlow Wrapper
* @param sentenceStartTokenId
* Id of sentence start Token
* @param sentenceEndTokenId
* Id of sentence end Token.
* @param configProtoBytes
* Configuration for TensorFlow session
*/
private[johnsnowlabs] class DistilBert(
val tensorflowWrapper: Option[TensorflowWrapper],
val onnxWrapper: Option[OnnxWrapper],
sentenceStartTokenId: Int,
sentenceEndTokenId: Int,
configProtoBytes: Option[Array[Byte]] = None,
signatures: Option[Map[String, String]] = None,
modelArch: String = ModelArch.wordEmbeddings)
extends Serializable {
protected val logger: Logger = LoggerFactory.getLogger("DistilBert")
val _tfBertSignatures: Map[String, String] = signatures.getOrElse(ModelSignatureManager.apply())
val detectedEngine: String =
if (tensorflowWrapper.isDefined) TensorFlow.name
else if (onnxWrapper.isDefined) ONNX.name
else TensorFlow.name
private val onnxSessionOptions: Map[String, String] = new OnnxSession().getSessionOptions
private def sessionWarmup(): Unit = {
val dummyInput =
Array(101, 2292, 1005, 1055, 4010, 6279, 1996, 5219, 2005, 1996, 2034, 28937, 1012, 102)
if (modelArch == ModelArch.wordEmbeddings) {
tag(Seq(dummyInput))
} else if (modelArch == ModelArch.sentenceEmbeddings) {
tagSequence(Seq(dummyInput))
}
}
sessionWarmup()
def tag(batch: Seq[Array[Int]]): Seq[Array[Array[Float]]] = {
val maxSentenceLength = batch.map(pieceIds => pieceIds.length).max
val batchLength = batch.length
val embeddings = detectedEngine match {
case ONNX.name =>
// [nb of encoded sentences , maxSentenceLength]
val (runner, env) = onnxWrapper.get.getSession(onnxSessionOptions)
val tokenTensors =
OnnxTensor.createTensor(env, batch.map(x => x.map(x => x.toLong)).toArray)
val maskTensors =
OnnxTensor.createTensor(
env,
batch.map(sentence => sentence.map(x => if (x == 0L) 0L else 1L)).toArray)
val inputs =
Map("input_ids" -> tokenTensors, "attention_mask" -> maskTensors).asJava
// TODO: A try without a catch or finally is equivalent to putting its body in a block; no exceptions are handled.
try {
val results = runner.run(inputs)
try {
val embeddings = results
.get("last_hidden_state")
.get()
.asInstanceOf[OnnxTensor]
.getFloatBuffer
.array()
embeddings
} finally if (results != null) results.close()
} catch {
case e: Exception =>
// Handle exceptions by logging or other means.
e.printStackTrace()
Array.empty[Float] // Return an empty array or appropriate error handling
} finally {
// Close tensors outside the try-catch to avoid repeated null checks.
// These resources are initialized before the try-catch, so they should be closed here.
tokenTensors.close()
maskTensors.close()
}
case _ =>
val tensors = new TensorResources()
val (tokenTensors, maskTensors, segmentTensors) =
PrepareEmbeddings.prepareBatchTensorsWithSegment(
tensors,
batch,
maxSentenceLength,
batchLength)
val runner = tensorflowWrapper.get
.getTFSessionWithSignature(
configProtoBytes = configProtoBytes,
savedSignatures = signatures,
initAllTables = false)
.runner
runner
.feed(
_tfBertSignatures.getOrElse(
ModelSignatureConstants.InputIds.key,
"missing_input_id_key"),
tokenTensors)
.feed(
_tfBertSignatures
.getOrElse(ModelSignatureConstants.AttentionMask.key, "missing_input_mask_key"),
maskTensors)
.fetch(
_tfBertSignatures
.getOrElse(
ModelSignatureConstants.LastHiddenState.key,
"missing_sequence_output_key"))
val outs = runner.run().asScala
val embeddings = TensorResources.extractFloats(outs.head)
tokenTensors.close()
maskTensors.close()
segmentTensors.close()
tensors.clearSession(outs)
tensors.clearTensors()
embeddings
}
PrepareEmbeddings.prepareBatchWordEmbeddings(
batch,
embeddings,
maxSentenceLength,
batchLength)
}
/** @param batch
* batches of sentences
* @return
* batches of vectors for each sentence
*/
def tagSequence(batch: Seq[Array[Int]]): Array[Array[Float]] = {
val maxSentenceLength = batch.map(pieceIds => pieceIds.length).max
val batchLength = batch.length
val tensors = new TensorResources()
val (tokenTensors, maskTensors) =
PrepareEmbeddings.prepareBatchTensors(
tensors = tensors,
batch = batch,
maxSentenceLength = maxSentenceLength,
batchLength = batchLength)
val runner = tensorflowWrapper.get
.getTFSessionWithSignature(
configProtoBytes = configProtoBytes,
savedSignatures = signatures,
initAllTables = false)
.runner
runner
.feed(
_tfBertSignatures.getOrElse(ModelSignatureConstants.InputIds.key, "missing_input_id_key"),
tokenTensors)
.feed(
_tfBertSignatures
.getOrElse(ModelSignatureConstants.AttentionMask.key, "missing_input_mask_key"),
maskTensors)
.fetch(_tfBertSignatures
.getOrElse(ModelSignatureConstants.PoolerOutput.key, "missing_pooled_output_key"))
val outs = runner.run().asScala
val embeddings = TensorResources.extractFloats(outs.head)
tokenTensors.close()
maskTensors.close()
tensors.clearSession(outs)
tensors.clearTensors()
val dim = embeddings.length / batchLength
embeddings.grouped(dim).toArray
}
def predict(
sentences: Seq[WordpieceTokenizedSentence],
originalTokenSentences: Seq[TokenizedSentence],
batchSize: Int,
maxSentenceLength: Int,
caseSensitive: Boolean): Seq[WordpieceEmbeddingsSentence] = {
/*Run embeddings calculation by batches*/
sentences.zipWithIndex
.grouped(batchSize)
.flatMap { batch =>
val encoded = PrepareEmbeddings.prepareBatchInputsWithPadding(
batch,
maxSentenceLength,
sentenceStartTokenId,
sentenceEndTokenId)
val vectors = tag(encoded)
/*Combine tokens and calculated embeddings*/
batch.zip(vectors).map { case (sentence, tokenVectors) =>
val tokenLength = sentence._1.tokens.length
/*All wordpiece embeddings*/
val tokenEmbeddings = tokenVectors.slice(1, tokenLength + 1)
val originalIndexedTokens = originalTokenSentences(sentence._2)
/*Word-level and span-level alignment with Tokenizer
https://github.com/google-research/bert#tokenization
### Input
orig_tokens = ["John", "Johanson", "'s", "house"]
labels = ["NNP", "NNP", "POS", "NN"]
# bert_tokens == ["[CLS]", "john", "johan", "##son", "'", "s", "house", "[SEP]"]
# orig_to_tok_map == [1, 2, 4, 6]*/
val tokensWithEmbeddings =
sentence._1.tokens.zip(tokenEmbeddings).flatMap { case (token, tokenEmbedding) =>
val tokenWithEmbeddings = TokenPieceEmbeddings(token, tokenEmbedding)
val originalTokensWithEmbeddings = originalIndexedTokens.indexedTokens
.find(p => p.begin == tokenWithEmbeddings.begin)
.map { token =>
val originalTokenWithEmbedding = TokenPieceEmbeddings(
TokenPiece(
wordpiece = tokenWithEmbeddings.wordpiece,
token = if (caseSensitive) token.token else token.token.toLowerCase(),
pieceId = tokenWithEmbeddings.pieceId,
isWordStart = tokenWithEmbeddings.isWordStart,
begin = token.begin,
end = token.end),
tokenEmbedding)
originalTokenWithEmbedding
}
originalTokensWithEmbeddings
}
WordpieceEmbeddingsSentence(tokensWithEmbeddings, originalIndexedTokens.sentenceIndex)
}
}
.toSeq
}
def predictSequence(
tokens: Seq[WordpieceTokenizedSentence],
sentences: Seq[Sentence],
batchSize: Int,
maxSentenceLength: Int): Seq[Annotation] = {
/*Run embeddings calculation by batches*/
tokens
.zip(sentences)
.zipWithIndex
.grouped(batchSize)
.flatMap { batch =>
val tokensBatch = batch.map(x => (x._1._1, x._2))
val sentencesBatch = batch.map(x => x._1._2)
val encoded = PrepareEmbeddings.prepareBatchInputsWithPadding(
tokensBatch,
maxSentenceLength,
sentenceStartTokenId,
sentenceEndTokenId)
val embeddings = tagSequence(encoded)
sentencesBatch.zip(embeddings).map { case (sentence, vectors) =>
Annotation(
annotatorType = AnnotatorType.SENTENCE_EMBEDDINGS,
begin = sentence.start,
end = sentence.end,
result = sentence.content,
metadata = Map(
"sentence" -> sentence.index.toString,
"token" -> sentence.content,
"pieceId" -> "-1",
"isWordStart" -> "true"),
embeddings = vectors)
}
}
.toSeq
}
}