{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"人工智能是一個正在迅速發展的領域，並日益成爲一種主流。要想跟上人工智能技術的發展，最好的方法就是學習深度學習中的新技術。以下這 8 個開源工具，可以幫助你輕鬆實現深度學習。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"1.TensorFlow"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"這個強大的工具包開發於 2015 年，被譽爲機器\/深度學習中最容易使用和部署的工具之一。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.tensorflow.org\/","title":null,"type":null},"content":[{"type":"text","text":"TensorFlow"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 最初是由谷歌大腦團隊創建的，用於處理其研究和生產目標，該項目構建了深度神經網絡來執行自然語言處理、圖像識別和翻譯等任務。由於它提供了大量的免費工具、庫和社區資源，現在它的使用甚至被 Uber、Twitter 和 eBay 等公司廣泛接受。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"它可用於 Python、"},{"type":"link","attrs":{"href":"https:\/\/www.haskell.org\/","title":null,"type":null},"content":[{"type":"text","text":"Haskell"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"、C++、Java 甚至是 JavaScript。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"2.Keras"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/keras.io\/","title":null,"type":null},"content":[{"type":"text","text":"Keras"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 由Python 編寫，最初發佈於 2015 年，是一種高級神經網絡 "},{"type":"link","attrs":{"href":"https:\/\/www.altexsoft.com\/blog\/engineering\/what-is-api-definition-types-specifications-documentation\/","title":null,"type":null},"content":[{"type":"text","text":"API"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，旨在簡化機器學習和深度學習，可以在 TensorFlow 或 Theano 之上部署。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"Keras 以其高度的廣泛性、模塊化、易用的特點，通過簡單的原型設計實現了快速的實驗，在 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Central_processing_unit","title":null,"type":null},"content":[{"type":"text","text":"CPU"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Graphics_processing_unit","title":null,"type":null},"content":[{"type":"text","text":"GPU"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 上高效運行，這對研究工作至關重要。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"3.PaddlePaddle"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"並行分佈式深度學習，又稱 "},{"type":"link","attrs":{"href":"https:\/\/github.com\/PaddlePaddle\/","title":null,"type":null},"content":[{"type":"text","text":"PaddlePaddle"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，"},{"type":"text","text":"具有易用、高效、靈活和可伸縮等特點"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這是百度研發的深度學習平臺，爲百度內部多項產品提供深度學習算法支持。"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"它在 2016 年向專業社區開源，具有深度學習的先進功能、端到端的開發工具包，受到製造業和農業部門的使用者青睞。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"4.PyTorch"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這個開源工具包"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"使用 Python 腳本語言，一般用於 "},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/cloud\/learn\/natural-language-processing#:~:text=Natural%20language%20processing%20(NLP","title":null,"type":null},"content":[{"type":"text","text":"自然語言處理"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"和"},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/topics\/computer-vision","title":null,"type":null},"content":[{"type":"text","text":"計算機視覺"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"。它具有強大的 GPU、內存使用效率和動態計算圖，這使得它在協助開發 "},{"type":"link","attrs":{"href":"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/how-dynamic-neural-networks-work.html;jsessionid=0dd555d070bf6f6fd6c6182ca137","title":null,"type":null},"content":[{"type":"text","text":"動態神經網絡"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"方面很受歡迎，並能夠根據用戶的要求建立圖形和可視化。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"5.Theano"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這個開源庫"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"發佈於 2007 年，使用 Python 編寫腳本，允許用戶定義、定製和評估數學表達式，從而使深度學習模型的形成變得容易。由於它是最古老的可自由使用的庫之一，它激發了深度學習的發展，導致了高效的"},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Computer_algebra","title":null,"type":null},"content":[{"type":"text","text":"符號微分"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"（symbolic differentiation）。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"6.Weka"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.cs.waikato.ac.nz\/ml\/weka\/","title":null,"type":null},"content":[{"type":"text","text":"Weka"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","text":" "},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"最初是作爲一個機器學習開源工具包開發的，它還具有深度學習包的功能，無論是在包管理、分類還是可視化方面，它都可以在一個簡單的界面使用中進行更復雜的分析。它是一個 Java 應用程序，但可以通過"},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/cloud\/learn\/api#:~:text=Application%20programming%20interfaces%2C%20or%20APIs,and%20functionality%20easily%20and%20securely.","title":null,"type":null},"content":[{"type":"text","text":"應用編程接口"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"用於 R 或 Python。因爲它最初是爲了一個教學平臺而開發的，一般只限於教學工具或者小項目。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"7.Caffe"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http:\/\/caffe.berkeleyvision.org\/","title":null,"type":null},"content":[{"type":"text","text":"Caffe"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"（Convolutional Architecture for Fast Feature Embedding，意即快速特徵嵌入的卷積架構）是一個開源的深度學習網絡，是爲速度、表達和模塊化設計的。它使用了 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/C%2B%2B","title":null,"type":null},"content":[{"type":"text","text":"C++"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，但也有一個 Python 接口。它由伯克利人工智能研究團隊於 2017 年開發併發布，它有一個精心設計的架構，良好的代碼編寫和快速的性能，可以快速實現工業部署。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"9.Torch"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http:\/\/torch.ch\/","title":null,"type":null},"content":[{"type":"text","text":"Torch"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","text":" "},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"最初是在 2002 年作爲機器學習庫開發和發佈的，它提供了一系列用於深度學習的算法，重點是 GPU，並提供 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/IOS","title":null,"type":null},"content":[{"type":"text","text":"iOS"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和 "},{"type":"link","attrs":{"href":"https:\/\/www.techopedia.com\/definition\/4219\/android-platform","title":null,"type":null},"content":[{"type":"text","text":"Android"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 平臺支持。它由腳本語言 "},{"type":"link","attrs":{"href":"https:\/\/www.lua.org\/about.html","title":null,"type":null},"content":[{"type":"text","text":"LUA"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和底層的 C 組件組成，使得它使用起來簡單、高效、快速。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" "}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"作者介紹："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Rancho Labs，科技愛好者自媒體，培養年輕人的思維，使他們在編程\/機器人技術\/人工智能領域站穩腳跟。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"原文鏈接："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"https:\/\/rancholabs.medium.com\/open-source-tools-for-deep-learning-f4d68c2a39cd"}]}]}
8個用於深度學習的開源工具
{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"人工智能是一個正在迅速發展的領域，並日益成爲一種主流。要想跟上人工智能技術的發展，最好的方法就是學習深度學習中的新技術。以下這 8 個開源工具，可以幫助你輕鬆實現深度學習。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"1.TensorFlow"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"這個強大的工具包開發於 2015 年，被譽爲機器\/深度學習中最容易使用和部署的工具之一。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.tensorflow.org\/","title":null,"type":null},"content":[{"type":"text","text":"TensorFlow"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 最初是由谷歌大腦團隊創建的，用於處理其研究和生產目標，該項目構建了深度神經網絡來執行自然語言處理、圖像識別和翻譯等任務。由於它提供了大量的免費工具、庫和社區資源，現在它的使用甚至被 Uber、Twitter 和 eBay 等公司廣泛接受。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"它可用於 Python、"},{"type":"link","attrs":{"href":"https:\/\/www.haskell.org\/","title":null,"type":null},"content":[{"type":"text","text":"Haskell"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"、C++、Java 甚至是 JavaScript。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"2.Keras"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/keras.io\/","title":null,"type":null},"content":[{"type":"text","text":"Keras"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 由Python 編寫，最初發佈於 2015 年，是一種高級神經網絡 "},{"type":"link","attrs":{"href":"https:\/\/www.altexsoft.com\/blog\/engineering\/what-is-api-definition-types-specifications-documentation\/","title":null,"type":null},"content":[{"type":"text","text":"API"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，旨在簡化機器學習和深度學習，可以在 TensorFlow 或 Theano 之上部署。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"Keras 以其高度的廣泛性、模塊化、易用的特點，通過簡單的原型設計實現了快速的實驗，在 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Central_processing_unit","title":null,"type":null},"content":[{"type":"text","text":"CPU"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Graphics_processing_unit","title":null,"type":null},"content":[{"type":"text","text":"GPU"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 上高效運行，這對研究工作至關重要。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"3.PaddlePaddle"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"並行分佈式深度學習，又稱 "},{"type":"link","attrs":{"href":"https:\/\/github.com\/PaddlePaddle\/","title":null,"type":null},"content":[{"type":"text","text":"PaddlePaddle"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，"},{"type":"text","text":"具有易用、高效、靈活和可伸縮等特點"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這是百度研發的深度學習平臺，爲百度內部多項產品提供深度學習算法支持。"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"它在 2016 年向專業社區開源，具有深度學習的先進功能、端到端的開發工具包，受到製造業和農業部門的使用者青睞。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"4.PyTorch"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這個開源工具包"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"使用 Python 腳本語言，一般用於 "},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/cloud\/learn\/natural-language-processing#:~:text=Natural%20language%20processing%20(NLP","title":null,"type":null},"content":[{"type":"text","text":"自然語言處理"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"和"},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/topics\/computer-vision","title":null,"type":null},"content":[{"type":"text","text":"計算機視覺"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"。它具有強大的 GPU、內存使用效率和動態計算圖，這使得它在協助開發 "},{"type":"link","attrs":{"href":"https:\/\/www.mathworks.com\/help\/deeplearning\/ug\/how-dynamic-neural-networks-work.html;jsessionid=0dd555d070bf6f6fd6c6182ca137","title":null,"type":null},"content":[{"type":"text","text":"動態神經網絡"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"方面很受歡迎，並能夠根據用戶的要求建立圖形和可視化。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"5.Theano"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"這個開源庫"},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"發佈於 2007 年，使用 Python 編寫腳本，允許用戶定義、定製和評估數學表達式，從而使深度學習模型的形成變得容易。由於它是最古老的可自由使用的庫之一，它激發了深度學習的發展，導致了高效的"},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/Computer_algebra","title":null,"type":null},"content":[{"type":"text","text":"符號微分"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"（symbolic differentiation）。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"6.Weka"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"https:\/\/www.cs.waikato.ac.nz\/ml\/weka\/","title":null,"type":null},"content":[{"type":"text","text":"Weka"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","text":" "},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"最初是作爲一個機器學習開源工具包開發的，它還具有深度學習包的功能，無論是在包管理、分類還是可視化方面，它都可以在一個簡單的界面使用中進行更復雜的分析。它是一個 Java 應用程序，但可以通過"},{"type":"link","attrs":{"href":"https:\/\/www.ibm.com\/cloud\/learn\/api#:~:text=Application%20programming%20interfaces%2C%20or%20APIs,and%20functionality%20easily%20and%20securely.","title":null,"type":null},"content":[{"type":"text","text":"應用編程接口"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"用於 R 或 Python。因爲它最初是爲了一個教學平臺而開發的，一般只限於教學工具或者小項目。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"7.Caffe"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http:\/\/caffe.berkeleyvision.org\/","title":null,"type":null},"content":[{"type":"text","text":"Caffe"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"（Convolutional Architecture for Fast Feature Embedding，意即快速特徵嵌入的卷積架構）是一個開源的深度學習網絡，是爲速度、表達和模塊化設計的。它使用了 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/C%2B%2B","title":null,"type":null},"content":[{"type":"text","text":"C++"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"，但也有一個 Python 接口。它由伯克利人工智能研究團隊於 2017 年開發併發布，它有一個精心設計的架構，良好的代碼編寫和快速的性能，可以快速實現工業部署。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"9.Torch"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"link","attrs":{"href":"http:\/\/torch.ch\/","title":null,"type":null},"content":[{"type":"text","text":"Torch"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","text":" "},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":"最初是在 2002 年作爲機器學習庫開發和發佈的，它提供了一系列用於深度學習的算法，重點是 GPU，並提供 "},{"type":"link","attrs":{"href":"https:\/\/en.wikipedia.org\/wiki\/IOS","title":null,"type":null},"content":[{"type":"text","text":"iOS"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和 "},{"type":"link","attrs":{"href":"https:\/\/www.techopedia.com\/definition\/4219\/android-platform","title":null,"type":null},"content":[{"type":"text","text":"Android"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 平臺支持。它由腳本語言 "},{"type":"link","attrs":{"href":"https:\/\/www.lua.org\/about.html","title":null,"type":null},"content":[{"type":"text","text":"LUA"}],"marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}]},{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" 和底層的 C 組件組成，使得它使用起來簡單、高效、快速。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"color","attrs":{"color":"#494949","name":"user"}}],"text":" "}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"作者介紹："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Rancho Labs，科技愛好者自媒體，培養年輕人的思維，使他們在編程\/機器人技術\/人工智能領域站穩腳跟。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"原文鏈接："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"https:\/\/rancholabs.medium.com\/open-source-tools-for-deep-learning-f4d68c2a39cd"}]}]}
發表評論
所有評論
還沒有人評論，想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.
相關文章