Spark NLP: State of the Art Natural Language Processing

Spark NLP is a Natural Language Processing library built on top of Apache Spark ML. It provides simple, performant & accurate NLP annotations for machine learning pipelines that scale easily in a distributed environment. Spark NLP comes with 160+ pretrained pipelines and models in more than 20+ languages. It supports state-of-the-art transformers such as BERT, XLNet, ELMO, ALBERT, and Universal Sentence Encoder that can be used seamlessly in a cluster. It also offers Tokenization, Part-of-Speech Tagging, Named Entity Recognition, Dependency Parsing, Spell Checking, Multi-class text classification, Multi-class sentiment analysis, and many more NLP tasks.

Project's website

Take a look at our official Spark NLP page: http://nlp.johnsnowlabs.com/ for user documentation and examples

Community support

• Slack For live discussion with the Spark NLP community and the team
• GitHub Bug reports, feature requests, and contributions
• Medium Spark NLP articles
• YouTube Spark NLP video tutorials

Table of contents

• Features
• Requirements
• Quick Start
• Apache Spark Support
• Databricks Support
• EMR Support
• Using Spark NLP
  • Spark Packages
  • Scala
    • Maven
    • SBT
  • Python
    • Pip/Conda
  • Compiled JARs
  • Apache Zeppelin
  • Jupyter Notebook
  • Google Colab Notebook
  • Databricks Cluser
  • S3 Cluster
• Pipelines & Models
  • Pipelines
  • Models
• Examples
• FAQ
• Troubleshooting
• Acknowledgments
• Contributing

Features

• Tokenization
• Stop Words Removal
• Normalizer
• Stemmer
• Lemmatizer
• NGrams
• Regex Matching
• Text Matching
• Chunking
• Date Matcher
• Sentence Detector
• Part-of-speech tagging
• Sentiment Detection (ML models)
• Spell Checker (ML and DL models)
• Word Embeddings (GloVe and Word2Vec)
• BERT Embeddings (TF Hub models)
• ELMO Embeddings (TF Hub models)
• ALBERT Embeddings (TF Hub models)
• XLNet Embeddings
• Universal Sentence Encoder (TF Hub models)
• Sentence Embeddings
• Chunk Embeddings
• Language Detection / Identification
• Multi-class Sentiment analysis (Deep learning)
• Multi-class Text Classification (Deep learning)
• Named entity recognition (Deep learning)
• Dependency parsing (Labeled/unlabled)
• Easy TensorFlow integration
• GPU Support
• Full integration with Spark ML functions
• +90 pre-trained models in 21 languages!
• +67 pre-trained pipelines in 10 languages!
• Multi-lingual NER models: Dutch, English, French, German, Italian, Norwegian, Polish, Portuguese, Russian, Spanish

Requirements

In order to use Spark NLP you need the following requirements:

• Java 8
• Apache Spark 2.4.x

Quick Start

This is a quick example of how to use Spark NLP pre-trained pipeline in Python and PySpark:

$ java -version
# should be Java 8 (Oracle or OpenJDK)
$ conda create -n sparknlp python=3.6 -y
$ conda activate sparknlp
$ pip install spark-nlp==2.5.2 pyspark==2.4.4

In Python console or Jupyter Python3 kernel:

# Import Spark NLP
from sparknlp.base import *
from sparknlp.annotator import *
from sparknlp.pretrained import PretrainedPipeline
import sparknlp

# Start Spark Session with Spark NLP
spark = sparknlp.start()

# Download a pre-trained pipeline
pipeline = PretrainedPipeline('explain_document_dl', lang='en')

# Your testing dataset
text = """
The Mona Lisa is a 16th century oil painting created by Leonardo.
It's held at the Louvre in Paris.
"""

# Annotate your testing dataset
result = pipeline.annotate(text)

# What's in the pipeline
list(result.keys())
Output: ['entities', 'stem', 'checked', 'lemma', 'document',
'pos', 'token', 'ner', 'embeddings', 'sentence']

# Check the results
result['entities']
Output: ['Mona Lisa', 'Leonardo', 'Louvre', 'Paris']

For more examples you can visit our dedicated repository to showcase all Spark NLP use cases!

Apache Spark Support

Spark NLP 2.5.2 has been built on top of Apache Spark 2.4.x

Find out more about Spark NLP versions from our release notes.

** Spark NLP is built and released based on Apache Spark 2.4.x, in order to use it with Apache Spark 2.3.x you need to manually compile it by changing the version in our build.sbt file.

** We do have the Fat JAR of Spark NLP 2.4.0 release already compiled for Apache Spark 2.3.x and it can be downloaded from our S3 from here.

** In case of using Apache Spark 2.3.x, the .pretrained() function won't download the models/pipelines automatically. You need to download them manyall and use .loat() instead.

Databricks Support

Spark NLP 2.5.2 has been tested and is compatible with the following runtimes:

• 6.2
• 6.2 ML
• 6.3
• 6.3 ML
• 6.4
• 6.4 ML
• 6.5
• 6.5 ML

EMR Support

Spark NLP 2.5.2 has been tsted and is compatible with the following EMR releases:

• 5.26.0
• 5.27.0

Full list of EMR releases.

Usage

Spark Packages

Command line (requires internet connection)

This library has been uploaded to the spark-packages repository.

Benefit of spark-packages is that makes it available for both Scala-Java and Python

To use the most recent version just add the --packages com.johnsnowlabs.nlp:spark-nlp_2.11: to you spark command

spark-shell --packages com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2

pyspark --packages com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2

spark-submit --packages com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2

This can also be used to create a SparkSession manually by using the spark.jars.packages option in both Python and Scala.

NOTE: To use SPark NLP with GPU you can use the dedicated GPU package com.johnsnowlabs.nlp:spark-nlp-gpu_2.11:2.5.2

Scala

Our package is deployed to maven central. In order to add this package as a dependency in your application:

Maven

spark-nlp:

<!-- https://mvnrepository.com/artifact/com.johnsnowlabs.nlp/spark-nlp -->
<dependency>
    <groupId>com.johnsnowlabs.nlp</groupId>
    <artifactId>spark-nlp_2.11</artifactId>
    <version>2.5.2</version>
</dependency>

spark-nlp-gpu:

<!-- https://mvnrepository.com/artifact/com.johnsnowlabs.nlp/spark-nlp-gpu -->
<dependency>
    <groupId>com.johnsnowlabs.nlp</groupId>
    <artifactId>spark-nlp-gpu_2.11</artifactId>
    <version>2.5.2</version>
</dependency>

SBT

spark-nlp:

// https://mvnrepository.com/artifact/com.johnsnowlabs.nlp/spark-nlp
libraryDependencies += "com.johnsnowlabs.nlp" %% "spark-nlp" % "2.5.2"

spark-nlp-gpu:

// https://mvnrepository.com/artifact/com.johnsnowlabs.nlp/spark-nlp-gpu
libraryDependencies += "com.johnsnowlabs.nlp" %% "spark-nlp-gpu" % "2.5.2"

Maven Central: https://mvnrepository.com/artifact/com.johnsnowlabs.nlp

Python

Python without explicit Pyspark installation

Pip/Conda

If you installed pyspark through pip/conda, you can install spark-nlp through the same channel.

Pip:

pip install spark-nlp==2.5.2

Conda:

conda install -c johnsnowlabs spark-nlp

PyPI spark-nlp package / Anaconda spark-nlp package

Then you'll have to create a SparkSession either from Spark NLP:

import sparknlp

spark = sparknlp.start()

or manually:

spark = SparkSession.builder \
    .appName("Spark NLP")\
    .master("local[4]")\
    .config("spark.driver.memory","16G")\
    .config("spark.driver.maxResultSize", "2G") \
    .config("spark.jars.packages", "com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2")\
    .config("spark.kryoserializer.buffer.max", "1000M")\
    .getOrCreate()

If using local jars, you can use spark.jars instead for a comma delimited jar files. For cluster setups, of course you'll have to put the jars in a reachable location for all driver and executor nodes.

Quick example:

import sparknlp
from sparknlp.pretrained import PretrainedPipeline

#create or get Spark Session

spark = sparknlp.start()

sparknlp.version()
spark.version

#download, load, and annotate a text by pre-trained pipeline

pipeline = PretrainedPipeline('recognize_entities_dl', 'en')
result = pipeline.annotate('The Mona Lisa is a 16th century oil painting created by Leonardo')

Compiled JARs

Build from source

spark-nlp

• FAT-JAR for CPU

sbt assembly

• FAT-JAR for GPU

sbt -Dis_gpu=true assembly

• Packaging the project

sbt package

Using the jar manually

If for some reason you need to use the JAR, you can either download the Fat JARs provided here or download it from Maven Central.

To add JARs to spark programs use the --jars option:

spark-shell --jars spark-nlp.jar

The preferred way to use the library when running spark programs is using the --packages option as specified in the spark-packages section.

Apache Zeppelin

Use either one of the following options

• Add the following Maven Coordinates to the interpreter's library list

com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2

• Add path to pre-built jar from here in the interpreter's library list making sure the jar is available to driver path

Python in Zeppelin

Apart from previous step, install python module through pip

pip install spark-nlp==2.5.2

Or you can install spark-nlp from inside Zeppelin by using Conda:

python.conda install -c johnsnowlabs spark-nlp

Configure Zeppelin properly, use cells with %spark.pyspark or any interpreter name you chose.

Finally, in Zeppelin interpreter settings, make sure you set properly zeppelin.python to the python you want to use and installed the pip library with (e.g. python3).

An alternative option would be to set SPARK_SUBMIT_OPTIONS (zeppelin-env.sh) and make sure --packages is there as shown earlier, since it includes both scala and python side installation.

Jupyter Notebook (Python)

Easiest way to get this done is by making Jupyter Notebook run using pyspark as follows:

export SPARK_HOME=/path/to/your/spark/folder
export PYSPARK_PYTHON=python3
export PYSPARK_DRIVER_PYTHON=jupyter
export PYSPARK_DRIVER_PYTHON_OPTS=notebook

pyspark --packages com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2

Alternatively, you can mix in using --jars option for pyspark + pip install spark-nlp

If not using pyspark at all, you'll have to run the instructions pointed here

Google Colab Notebook

Google Colab is perhaps the easiest way to get started with spark-nlp. It requires no installation or set up other than having a Google account.

Run the following code in Google Colab notebook and start using spark-nlp right away.

import os

# Install java
! apt-get install -y openjdk-8-jdk-headless -qq > /dev/null
os.environ["JAVA_HOME"] = "/usr/lib/jvm/java-8-openjdk-amd64"
os.environ["PATH"] = os.environ["JAVA_HOME"] + "/bin:" + os.environ["PATH"]
! java -version

# Install pyspark
! pip install --ignore-installed pyspark==2.4.4

# Install Spark NLP
! pip install --ignore-installed spark-nlp==2.5.2

# Quick SparkSession start
import sparknlp
spark = sparknlp.start()

print("Spark NLP version")
sparknlp.version()
print("Apache Spark version")
spark.version

Here is a live demo on Google Colab that performs sentiment analysis and NER using pretrained spark-nlp models.

Databricks Cluster

1. Create a cluster if you don't have one already

2. On a new cluster or existing one you need to add the following to the Aadvanced Options -> Spark tab:

spark.kryoserializer.buffer.max 1000M
spark.serializer org.apache.spark.serializer.KryoSerializer

3. In Libraries tab inside your cluster you need to follow these steps:

   3.1. Insatll New -> PyPI -> spark-nlp -> Install

   3.2. Install New -> Maven -> Coordinates -> com.johnsnowlabs.nlp:spark-nlp_2.11:2.5.2 -> Install

4. Now you can attach your notebook to the cluster and use Spark NLP!

S3 Cluster

With no hadoop configuration

If your distributed storage is S3 and you don't have a standard hadoop configuration (i.e. fs.defaultFS) You need to specify where in the cluster distributed storage you want to store Spark NLP's tmp files. First, decide where you want to put your application.conf file

import com.johnsnowlabs.util.ConfigLoader
ConfigLoader.setConfigPath("/somewhere/to/put/application.conf")

And then we need to put in such application.conf the following content

sparknlp {
  settings {
    cluster_tmp_dir = "somewhere in s3n:// path to some folder"
  }
}

Pipelines and Models

Pipelines

Spark NLP offers more than 30 pre-trained pipelines in 6 languages.

English pipelines:

Quick example:

import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
import com.johnsnowlabs.nlp.SparkNLP

SparkNLP.version()

val testData = spark.createDataFrame(Seq(
(1, "Google has announced the release of a beta version of the popular TensorFlow machine learning library"),
(2, "Donald John Trump (born June 14, 1946) is the 45th and current president of the United States")
)).toDF("id", "text")

val pipeline = PretrainedPipeline("explain_document_dl", lang="en")

val annotation = pipeline.transform(testData)

annotation.show()
/*
import com.johnsnowlabs.nlp.pretrained.PretrainedPipeline
import com.johnsnowlabs.nlp.SparkNLP
2.5.0
testData: org.apache.spark.sql.DataFrame = [id: int, text: string]
pipeline: com.johnsnowlabs.nlp.pretrained.PretrainedPipeline = PretrainedPipeline(explain_document_dl,en,public/models)
annotation: org.apache.spark.sql.DataFrame = [id: int, text: string ... 10 more fields]
+---+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+
| id|                text|            document|               token|            sentence|             checked|               lemma|                stem|                 pos|          embeddings|                 ner|            entities|
+---+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+
|  1|Google has announ...|[[document, 0, 10...|[[token, 0, 5, Go...|[[document, 0, 10...|[[token, 0, 5, Go...|[[token, 0, 5, Go...|[[token, 0, 5, go...|[[pos, 0, 5, NNP,...|[[word_embeddings...|[[named_entity, 0...|[[chunk, 0, 5, Go...|
|  2|The Paris metro w...|[[document, 0, 11...|[[token, 0, 2, Th...|[[document, 0, 11...|[[token, 0, 2, Th...|[[token, 0, 2, Th...|[[token, 0, 2, th...|[[pos, 0, 2, DT, ...|[[word_embeddings...|[[named_entity, 0...|[[chunk, 4, 8, Pa...|
+---+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+
*/

annotation.select("entities.result").show(false)

/*
+----------------------------------+
|result                            |
+----------------------------------+
|[Google, TensorFlow]              |
|[Donald John Trump, United States]|
+----------------------------------+
*/

Please check our dedicated repository for the full list of pre-trained pipelines

Models

Spark NLP offers more than 90 pre-trained models in 21 languages.

English Models:

Embeddings:

Classification:

Quick online example:

# load NER model trained by deep learning approach and GloVe word embeddings
ner_dl = NerDLModel.pretrained('ner_dl')
# load NER model trained by deep learning approach and BERT word embeddings
ner_bert = NerDLModel.pretrained('ner_dl_bert')

// load French POS tagger model trained by Universal Dependencies
val french_pos = PerceptronModel.pretrained("pos_ud_gsd", lang="fr")
// load Italain LemmatizerModel
val italian_lemma = LemmatizerModel.pretrained("lemma_dxc", lang="it")

Quick offline example:

• Loading PerceptronModel annotator model inside Spark NLP Pipeline

val french_pos = PerceptronModel.load("/tmp/pos_ud_gsd_fr_2.0.2_2.4_1556531457346/")
      .setInputCols("document", "token")
      .setOutputCol("pos")

Please check our dedicated repository for the full list of pre-trained models

Examples

Need more examples? Check out our dedicated repository to showcase all Spark NLP use cases!

All examples: spark-nlp-workshop

FAQ

Check our Articles and FAQ page here

Acknowledgments

Special community acknowledgments

Thanks in general to the community who have been lately reporting important issues and pull request with bugfixes. Community has been key in the last releases with feedback in various Spark based environments.

Here a few specific mentions for recurring feedback and slack participation

• @maziyarpanahi - For contributing with testing and valuable feedback
• @easimadi - For contributing with documentation and valuable feedback

Contributing

We appreciate any sort of contributions:

• ideas
• feedback
• documentation
• bug reports
• nlp training and testing corpora
• development and testing

Clone the repo and submit your pull-requests! Or directly create issues in this repo.

Contact

nlp@johnsnowlabs.com

John Snow Labs

http://johnsnowlabs.com