Deeplearning4j

Eclipse Deeplearning4j is a programming library

Eclipse Deeplearning4j
written in Java for the Java virtual machine
(JVM).[2][3] It is a framework with wide support for Original author(s) Alex D. Black, Adam
deep learning algorithms.[4] Deeplearning4j includes Gibson, Vyacheslav
implementations of the restricted Boltzmann machine, Kokorin, Josh Patterson
deep belief net, deep autoencoder, stacked denoising Developer(s) Kondiut K. K. and
autoencoder and recursive neural tensor network, contributors
word2vec, doc2vec, and GloVe. These algorithms all Preview release 1.0.0-beta7 / 13 May
include distributed parallel versions that integrate with 2020[1]
Apache Hadoop and Spark.[5]
Repository github.com/deeplearning4j
Deeplearning4j is open-source software released under /deeplearning4j (https://gith
Apache License 2.0,[6] developed mainly by a machine ub.com/deeplearning4j/dee
learning group headquartered in San Francisco.[7] It is plearning4j)
supported commercially by the startup Skymind, which Written in Java, CUDA, C, C++,
bundles DL4J, TensorFlow, Keras and other deep Operating system Linux, macOS, Windows,
learning libraries in an enterprise distribution called the Android, iOS
Skymind Intelligence Layer.[8] Deeplearning4j was
Platform CUDA, x86, ARM,
contributed to the Eclipse Foundation in October
PowerPC
2017.[9][10]
Available in English
Type Natural language
Introduction processing, deep learning,
machine vision, artificial
Deeplearning4j relies on the widely used programming intelligence
language Java, though it is compatible with Clojure License Apache License 2.0
and includes a Scala application programming
Website www.deeplearning4j.org (h
interface (API). It is powered by its own open-source
ttp://www.deeplearning4j.or
numerical computing library, ND4J, and works with
g)
both central processing units (CPUs) and graphics
processing units (GPUs).[11][12]

Deeplearning4j has been used in several commercial and academic applications. The code is hosted on
GitHub.[13] A support forum is maintained on Gitter.[14]

The framework is composable, meaning shallow neural nets such as restricted Boltzmann machines,
convolutional nets, autoencoders, and recurrent nets can be added to one another to create deep nets of
varying types. It also has extensive visualization tools,[15] and a computation graph.[16]

Distributed
Training with Deeplearning4j occurs in a cluster. Neural nets are trained in parallel via iterative reduce,
which works on Hadoop-YARN and on Spark.[7][17] Deeplearning4j also integrates with CUDA kernels
to conduct pure GPU operations, and works with distributed GPUs.

Scientific computing for the JVM

Deeplearning4j includes an n-dimensional array class using ND4J that allows scientific computing in
Java and Scala, similar to the functions that NumPy provides to Python. It's effectively based on a library
for linear algebra and matrix manipulation in a production environment.

DataVec vectorization library for machine-learning

DataVec vectorizes various file formats and data types using an input/output format system similar to
Hadoop's use of MapReduce; that is, it turns various data types into columns of scalars termed vectors.
DataVec is designed to vectorize CSVs, images, sound, text, video, and time series.[18][19]

Text and NLP

Deeplearning4j includes a vector space modeling and topic modeling toolkit, implemented in Java and
integrating with parallel GPUs for performance. It is designed to handle large text sets.

Deeplearning4j includes implementations of term frequency–inverse document frequency (tf–idf), deep

learning, and Mikolov's word2vec algorithm,[20] doc2vec, and GloVe, reimplemented and optimized in
Java. It relies on t-distributed stochastic neighbor embedding (t-SNE) for word-cloud visualizations.

Real-world use cases and integrations

Real-world use cases for Deeplearning4j include network intrusion detection and cybersecurity, fraud
detection for the financial sector,[21][22] anomaly detection in industries such as manufacturing,
recommender systems in e-commerce and advertising,[23] and image recognition.[24] Deeplearning4j has
integrated with other machine-learning platforms such as RapidMiner, Prediction.io,[25] and Weka.[26]

Machine Learning Model Server

Deeplearning4j serves machine-learning models for inference in production using the free developer
edition of SKIL, the Skymind Intelligence Layer.[27][28] A model server serves the parametric machine-
learning models that makes decisions about data. It is used for the inference stage of a machine-learning
workflow, after data pipelines and model training. A model server is the tool that allows data science
research to be deployed in a real-world production environment.
What a Web server is to the Internet, a model server is to AI. Where a Web server receives an HTTP
request and returns data about a Web site, a model server receives data, and returns a decision or
prediction about that data: e.g. sent an image, a model server might return a label for that image,
identifying faces or animals in photographs.

The SKIL model server is able to import models from Python frameworks such as Tensorflow, Keras,
Theano and CNTK, overcoming a major barrier in deploying deep learning models.

Benchmarks
Deeplearning4j is as fast as Caffe for non-trivial image recognition tasks using multiple GPUs.[29] For
programmers unfamiliar with HPC on the JVM, there are several parameters that must be adjusted to
optimize neural network training time. These include setting the heap space, the garbage collection
algorithm, employing off-heap memory and pre-saving data (pickling) for faster ETL.[30] Together, these
optimizations can lead to a 10x acceleration in performance with Deeplearning4j.

API Languages: Java, Scala, Python, Clojure & Kotlin

Deeplearning4j can be used via multiple API languages including Java, Scala, Python, Clojure and
Kotlin. Its Scala API is called ScalNet.[31] Keras serves as its Python API.[32] And its Clojure wrapper is
known as DL4CLJ.[33] The core languages performing the large-scale mathematical operations necessary
for deep learning are C, C++ and CUDA C.

Tensorflow, Keras & Deeplearning4j

Tensorflow, Keras and Deeplearning4j work together. Deeplearning4j can import models from
Tensorflow and other Python frameworks if they have been created with Keras.[34]

See also
Free and open-
source software
portal
Computer
programming portal

Comparison of deep learning software

Artificial intelligence
Machine learning
Deep learning

References
1. "Releases · eclipse/deeplearning4j" (https://github.com/eclipse/deeplearning4j/releases).
 github.com. Retrieved 2021-04-03.
2. Metz, Cade (2014-06-02). "The Mission to Bring Google's AI to the Rest of the World" (http
s://www.wired.com/2014/06/skymind-deep-learning/). Wired.com. Retrieved 2014-06-28.
3. Vance, Ashlee (2014-06-03). "Deep Learning for (Some of) the People" (https://web.archive.
org/web/20140604063205/http://www.businessweek.com/articles/2014-06-03/teaching-small
er-companies-how-to-probe-deep-learning-on-their-own). Bloomberg Businessweek.
Archived from the original (http://www.businessweek.com/articles/2014-06-03/teaching-smal
ler-companies-how-to-probe-deep-learning-on-their-own) on June 4, 2014. Retrieved
2014-06-28.
4. Novet, Jordan (2015-11-14). "Want an open-source deep learning framework? Take your
pick" (https://venturebeat.com/2015/11/14/deep-learning-frameworks/). VentureBeat.
Retrieved 2015-11-24.
5. "Adam Gibson, DeepLearning4j on Spark and Data Science on JVM with nd4j, SF Spark
@Galvanize 20150212" (https://www.youtube.com/watch?v=LCsc1hFuNac). SF Spark
Meetup. 2015-02-12. Retrieved 2015-03-01.
6. "Github Repository" (https://github.com/agibsonccc/java-deeplearning). GitHub. April 2020.
7. "deeplearning4j.org" (http://deeplearning4j.org/).
8. "Skymind Intelligence Layer Community Edition" (https://web.archive.org/web/20171107015
537/https://skymind.ai/quickstart). Archived from the original (https://skymind.ai/quickstart)
on 2017-11-07. Retrieved 2017-11-02.
9. "Eclipse Deeplearning4j Project Page" (https://projects.eclipse.org/proposals/deeplearning4
j). 22 June 2017.
10. "Skymind's Deeplearning4j, the Eclipse Foundation, and scientific computing in the JVM" (ht
tps://jaxenter.com/skymind-deeplearning4j-eclipse-138872.html). Jaxenter. 13 November
2017. Retrieved 2017-11-15.
11. Novet, Jordan (September 28, 2016). "Deep learning startup Skymind raises $3 million,
launches Intelligence Layer distribution" (https://venturebeat.com/entrepreneur/deep-learnin
g-startup-skymind-raises-3-million-launches-intelligence-layer-distribution/). VentureBeat.
12. Novet, Jordan (June 2, 2014). "Skymind launches with open-source, plug-and-play deep
learning features for your app" (https://venturebeat.com/dev/skymind-launches-with-open-so
urce-plug-and-play-deep-learning-features-for-your-app/). VentureBeat.
13. "deeplearning4j/deeplearning4j" (https://github.com/deeplearning4j/deeplearning4j). 29 April
2023. Retrieved 29 April 2023 – via GitHub.
14. "Element" (https://app.gitter.im/#/room/#deeplearning4j_deeplearning4j:gitter.im).
app.gitter.im. Retrieved 29 April 2023.
15. "Deeplearning4j Visualization Tools" (https://web.archive.org/web/20170810053823/https://d
eeplearning4j.org/visualization). Archived from the original (http://deeplearning4j.org/visualiz
ation) on 2017-08-10. Retrieved 2016-08-17.
16. "Deeplearning4j Computation Graph" (https://web.archive.org/web/20170810053821/https://
deeplearning4j.org/compgraph). Archived from the original (http://deeplearning4j.org/compgr
aph) on 2017-08-10. Retrieved 2016-08-17.
17. "Iterative reduce" (https://github.com/emsixteeen/IterativeReduce). GitHub. 15 March 2020.
18. "DataVec ETL for Machine Learning" (https://web.archive.org/web/20171002011151/https://
deeplearning4j.org/datavec). Archived from the original (http://deeplearning4j.org/datavec)
on 2017-10-02. Retrieved 2016-09-18.
19. "Anomaly Detection for Time Series Data with Deep Learning" (https://www.infoq.com/article
s/deep-learning-time-series-anomaly-detection/). InfoQ. Retrieved 29 April 2023.
20. "Google Code Archive - Long-term storage for Google Code Project Hosting" (https://code.g
oogle.com/archive/p/word2vec). code.google.com. Retrieved 29 April 2023.
21. "Archived copy" (https://web.archive.org/web/20160310082208/http://www.skymind.io/financ
e/). Archived from the original (http://www.skymind.io/finance/) on 2016-03-10. Retrieved
2016-02-22.
22. "skymind.ai" (https://skymind.ai/bsa-aml%7B%7BDead+link%7Cdate=September+2018+%7
Cbot=InternetArchiveBot+%7Cfix-attempted=yes+%7D%7D). skymind.ai. Retrieved 29 April
2023.
23. "Archived copy" (https://web.archive.org/web/20160310082156/http://www.skymind.io/comm
erce/). Archived from the original (http://www.skymind.io/commerce/) on 2016-03-10.
Retrieved 2016-02-22.
24. "skymind.ai" (https://skymind.ai/image%7B%7BDead+link%7Cdate=September+2018+%7C
bot=InternetArchiveBot+%7Cfix-attempted=yes+%7D%7D). skymind.ai. Retrieved 29 April
2023.
25. "DeepLearning4J(Stable) | RapidMiner China" (https://archive.today/20160518041312/http
s://www.rapidminerchina.com/en/products/shop/product/deeplearning4j/).
www.rapidminerchina.com. Archived from the original (https://www.rapidminerchina.com/en/
products/shop/product/deeplearning4j/) on 18 May 2016. Retrieved 22 May 2022.
26. "Home - WekaDeeplearning4j" (https://deeplearning.cms.waikato.ac.nz/).
deeplearning.cms.waikato.ac.nz. Retrieved 29 April 2023.
27. "Products" (https://web.archive.org/web/20170921001159/https://skymind.ai/products).
Archived from the original (https://skymind.ai/products) on 2017-09-21. Retrieved
2017-09-20.
28. "Model Server for Deep Learning and AI - Deeplearning4j: Open-source, Distributed Deep
Learning for the JVM" (https://web.archive.org/web/20170921001516/https://deeplearning4j.
org/modelserver). Archived from the original (https://deeplearning4j.org/modelserver) on
2017-09-21. Retrieved 2017-09-20.
29. "GitHub - deeplearning4j/Dl4j-benchmark: Repo to track dl4j benchmark code" (https://githu
b.com/deeplearning4j/dl4j-benchmark). GitHub. 19 December 2019.
30. "Deeplearning4j Benchmarks - Deeplearning4j: Open-source, Distributed Deep Learning for
the JVM" (https://web.archive.org/web/20170809124911/https://deeplearning4j.org/benchma
rk). Archived from the original (https://deeplearning4j.org/benchmark) on 2017-08-09.
Retrieved 2017-01-30.
31. "Scala, Spark and Deeplearning4j - Deeplearning4j: Open-source, Distributed Deep
Learning for the JVM" (https://web.archive.org/web/20170225133003/https://deeplearning4j.
org/scala). Archived from the original (https://deeplearning4j.org/scala) on 2017-02-25.
Retrieved 2017-02-25.
32. "Running Keras with Deeplearning4j - Deeplearning4j: Open-source, Distributed Deep
Learning for the JVM" (https://web.archive.org/web/20170225133010/https://deeplearning4j.
org/keras). Archived from the original (https://deeplearning4j.org/keras) on 2017-02-25.
Retrieved 2017-02-25.
33. "Deep Learning with Clojure - Deeplearning4j: Open-source, Distributed Deep Learning for
the JVM" (https://web.archive.org/web/20170225133007/https://deeplearning4j.org/clojure).
Archived from the original (https://deeplearning4j.org/clojure) on 2017-02-25. Retrieved
2017-02-25.
34. "Tensorflow & Deeplearning4j - Deeplearning4j: Open-source, Distributed Deep Learning for
the JVM" (https://web.archive.org/web/20170908021856/https://deeplearning4j.org/tensorflo
w). Archived from the original (https://deeplearning4j.org/tensorflow) on 2017-09-08.
Retrieved 2017-09-07.

Retrieved from "https://en.wikipedia.org/w/index.php?title=Deeplearning4j&oldid=1241720012"