欢迎大家赞助一杯啤酒🍺 我们准备了下酒菜：Formal mathematics/Isabelle/ML, Formal verification/Coq/ACL2, C++/F#/Lisp
Deeplearning4j
来自开放百科 - 灰狐
（版本间的差异）
小 (→图集) |
小 |
||
| （未显示1个用户的19个中间版本） | |||
| 第1行： | 第1行： | ||
| + | deeplearning4j（DL4j） | ||
| + | |||
| + | [[文件:DL4J-logo.png|right|DL4J]] | ||
| + | |||
| + | ==简介== | ||
| + | Eclipse Deeplearning4j | ||
| + | |||
deeplearning4j：开源(Apache v2)[[Artificial neural network|神经网络]]平台，基于 [[Apache Hadoop]], [[Apache Spark]] 构建，为 [[Java]], [[Scala]] & [[Clojure]] 编程语言提供的[[deep learning|深度学习]]库。 | deeplearning4j：开源(Apache v2)[[Artificial neural network|神经网络]]平台，基于 [[Apache Hadoop]], [[Apache Spark]] 构建，为 [[Java]], [[Scala]] & [[Clojure]] 编程语言提供的[[deep learning|深度学习]]库。 | ||
| + | |||
| + | |||
| + | ==新闻== | ||
| + | *2017 年 10 月，Skymind 加入了 Eclipse 基金会，并且将 DL4J 贡献给开源 Java Enterprise Edition 生态系统。 | ||
| + | |||
| + | ==指南== | ||
| + | *[https://developer.ibm.com/zh/technologies/artificial-intelligence/articles/cc-get-started-deeplearning4j/ Deeplearning4j 入门] | ||
| + | |||
| + | git clone https://github.com/eclipse/deeplearning4j-examples | ||
| + | cd deeplearning4j-examples | ||
| + | mvn clean install | ||
| + | 使用 [[IntelliJ IDEA]] 打开项目，[http://deeplearning4j.org/quickstart.html 完成DL4J第一次体验]。 | ||
==应用场景== | ==应用场景== | ||
| 第13行： | 第32行： | ||
==DL4J[http://deeplearning4j.org/neuralnet-overview 神经网络]== | ==DL4J[http://deeplearning4j.org/neuralnet-overview 神经网络]== | ||
[http://deeplearning4j.org/neuralnetworktable.html 如何选择神经网络] | [http://deeplearning4j.org/neuralnetworktable.html 如何选择神经网络] | ||
| − | *[ | + | *[https://deeplearning4j.org/cn/restrictedboltzmannmachine 受限玻尔兹曼机] |
| − | *[ | + | *[https://deeplearning4j.org/cn/convolutionalnets 卷积网络]（图像） |
| − | *[ | + | *[https://deeplearning4j.org/cn/lstm 递归网络/LSTMs]（时间序列和传感器数据） |
| − | *[https:// | + | *[https://deeplearning4j.org/cn/deepbeliefnetwork 深度置信网络] |
| − | *[ | + | *[https://deeplearning4j.org/cn/deepautoencoder 深度自动编码器]（问答、数据压缩） |
| − | *[ | + | *[https://deeplearning4j.org/cn/usingrnns 递归神经传感器网络]（场景、分析） |
| − | + | *[https://deeplearning4j.org/cn/stackeddenoisingautoencoder 堆叠式降噪自动编码器] | |
| − | *[ | + | *[https://github.com/deeplearning4j/rl4j 深度增强学习] |
==为何选择DL4J== | ==为何选择DL4J== | ||
| 第30行： | 第49行： | ||
*[[ND4J]]：线性代数库，较[[Numpy]]快一倍 | *[[ND4J]]：线性代数库，较[[Numpy]]快一倍 | ||
*[http://deeplearning4j.org/documentation 文档全面、有深度、多语言] | *[http://deeplearning4j.org/documentation 文档全面、有深度、多语言] | ||
| − | |||
| − | |||
==推荐引擎== | ==推荐引擎== | ||
| 第37行： | 第54行： | ||
==[[Apache Spark|Spark]]== | ==[[Apache Spark|Spark]]== | ||
| − | *[ | + | *[https://deeplearning4j.org/cn/spark 基于Spark的Deeplearning4j] |
| − | *[ | + | *[https://deeplearning4j.org/cn/iterativereduce DL4J与基于Hadoop和Spark的迭代式归纳] |
*[http://deeplearning4j.org/spark-fast-native-binaries.html How to Speed Up Spark With Native Binaries and OpenBlas] | *[http://deeplearning4j.org/spark-fast-native-binaries.html How to Speed Up Spark With Native Binaries and OpenBlas] | ||
| + | |||
| + | ==Kotlin== | ||
| + | [https://deeplearning4j.org/kotlin Using Deeplearning4j With Kotlin] | ||
| + | |||
| + | ==Scala== | ||
| + | [https://deeplearning4j.org/scala Scala, Apache Spark and Deeplearning4j] | ||
| + | *[https://github.com/deeplearning4j/ScalNet ScalNet]是受[[Keras]]启发而为Deeplearning4j开发的[[Scala]]语言封装 | ||
| + | *[https://github.com/deeplearning4j/nd4s ND4S] Scala bindings for [[ND4J]] | ||
| + | |||
| + | ==Clojure== | ||
| + | [https://deeplearning4j.org/clojure Deep Learning With Clojure] | ||
| + | |||
| + | ==[[C++]]== | ||
| + | *[[JavaCPP]] | ||
==图集== | ==图集== | ||
<gallery> | <gallery> | ||
image:deeplearning4j-overview.png|DL4J | image:deeplearning4j-overview.png|DL4J | ||
| + | image:dl4j-eco-cn.jpg|DL4J组件 | ||
image:deep-learning-use-case-industries.png|工业应用 | image:deep-learning-use-case-industries.png|工业应用 | ||
image:deeplearning4j-ui.png|DL4J UI | image:deeplearning4j-ui.png|DL4J UI | ||
| + | image:deeplearning4j-training-ui.png|训练UI | ||
image:neural-network-table.png|神经网络 | image:neural-network-table.png|神经网络 | ||
image:Deeplearning4j-Reference-Architecture-CPU-GPU-Training.png|DL4J参考架构:训练 | image:Deeplearning4j-Reference-Architecture-CPU-GPU-Training.png|DL4J参考架构:训练 | ||
image:Deeplearning4j-Reference-Architecture-CPU-GPU-Scoring.png|DL4J参加架构:得分 | image:Deeplearning4j-Reference-Architecture-CPU-GPU-Scoring.png|DL4J参加架构:得分 | ||
| + | image:SKIL-Reference-Architecture.png|SKIL参考架构 | ||
</gallery> | </gallery> | ||
==链接== | ==链接== | ||
| − | *[ | + | *[https://deeplearning4j.org/ deeplearning4j官网] |
| − | *[https://github.com/ | + | *[https://github.com/eclipse/deeplearning4j deeplearning4j @ github] |
*[http://docs.huihoo.com/javadoc/deeplearning4j/ deeplearning4j javadoc] | *[http://docs.huihoo.com/javadoc/deeplearning4j/ deeplearning4j javadoc] | ||
*[http://deeplearning4j.org/glossary 深度学习词汇表] | *[http://deeplearning4j.org/glossary 深度学习词汇表] | ||
*[http://deeplearning4j.org/compare-dl4j-torch7-pylearn.html DL4J vs. Torch vs. Theano vs. Caffe vs. TensorFlow] | *[http://deeplearning4j.org/compare-dl4j-torch7-pylearn.html DL4J vs. Torch vs. Theano vs. Caffe vs. TensorFlow] | ||
| − | *[ | + | *[https://skymind.ai Skymind]是DL4J的商业支持机构 |
[[category:deep learning]] | [[category:deep learning]] | ||
| 第67行： | 第101行： | ||
[[category:scala]] | [[category:scala]] | ||
[[category:clojure]] | [[category:clojure]] | ||
| + | [[category:eclipse]] | ||
| + | [[category:huihoo]] | ||
2022年4月11日 (一) 02:09的最后版本
deeplearning4j（DL4j）
目录 |
[编辑] 简介
Eclipse Deeplearning4j
deeplearning4j：开源(Apache v2)神经网络平台，基于 Apache Hadoop, Apache Spark 构建，为 Java, Scala & Clojure 编程语言提供的深度学习库。
[编辑] 新闻
- 2017 年 10 月，Skymind 加入了 Eclipse 基金会，并且将 DL4J 贡献给开源 Java Enterprise Edition 生态系统。
[编辑] 指南
git clone https://github.com/eclipse/deeplearning4j-examples cd deeplearning4j-examples mvn clean install
使用 IntelliJ IDEA 打开项目，完成DL4J第一次体验。
[编辑] 应用场景
神经网络应用情景
[编辑] DL4J神经网络
- 受限玻尔兹曼机
- 卷积网络（图像）
- 递归网络/LSTMs（时间序列和传感器数据）
- 深度置信网络
- 深度自动编码器（问答、数据压缩）
- 递归神经传感器网络（场景、分析）
- 堆叠式降噪自动编码器
- 深度增强学习
[编辑] 为何选择DL4J
为何选择Deeplearning4j？
- 功能多样的N维数组类，为Java和Scala设计
- 与GPU集合
- 可在Hadoop、Spark上实现伸缩
- Canova：机器学习库的通用向量化工具
- ND4J：线性代数库，较Numpy快一倍
- 文档全面、有深度、多语言
[编辑] 推荐引擎
Build a Recommendation Engine With DL4J
[编辑] Spark
- 基于Spark的Deeplearning4j
- DL4J与基于Hadoop和Spark的迭代式归纳
- How to Speed Up Spark With Native Binaries and OpenBlas
[编辑] Kotlin
Using Deeplearning4j With Kotlin
[编辑] Scala
Scala, Apache Spark and Deeplearning4j
[编辑] Clojure
[编辑] C++
[编辑] 图集
DL4J
DL4J组件
工业应用
DL4J UI
训练UI
神经网络
DL4J参考架构:训练
DL4J参加架构:得分
SKIL参考架构
[编辑] 链接
分享您的观点
