Jump to content

TensorFlow

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Yobot (talk | contribs) at 15:49, 3 December 2016 (→TensorFlow: WP:CHECKWIKI error fixes (#104) using AWB). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

TensorFlow
Developer(s)Google Brain Team[1]
Initial releaseNovember 9, 2015; 9 years ago (2015-11-09)
Stable release
0.11.0[2]
Repository
Written inPython, C++
PlatformLinux, Mac OS X, Windows
TypeMachine Learning Library
LicenseApache 2.0 open source license
Websitewww.tensorflow.org

TensorFlow is an open source software library for machine learning in various kinds of perceptual and language understanding tasks.[3] It is currently used for both research and production by 50[3]: min 0:15/2:17  different teams in dozens[4]: p.2  of commercial Google products, such as speech recognition, Gmail, Google Photos, and search,[3]: 0:26/2:17  many of which had previously used its predecessor DistBelief. TensorFlow was originally developed by the Google Brain team for Google's research and production purposes and later released under the Apache 2.0 open source license on November 9, 2015.[1][5]

History

DistBelief

Starting in 2011, Google Brain built DistBelief as their first-generation, proprietary, machine learning system. More than 50 teams at Google and other Alphabet companies have deployed DistBelief's deep learning neural networks in Google's commercial products, including Google Search, Google Voice Search, advertising, Google Photos, Google Maps, Google Street View, Google Translate, and YouTube.[4][6] Google assigned computer scientists, such as Geoffrey Hinton and Jeff Dean, to simplify and refactor the codebase of DistBelief into a faster, more robust application-grade library, which became TensorFlow.[7] In 2009, the team led by Hinton was able to substantially reduce the number of errors in neural networks which used DistBelief; this breakthrough was made possible by Hinton's scientific breakthroughs in generalized backpropagation. Most notably, Hinton's breakthrough directly led to reducing errors in Google’s speech recognition software by at least 25 percent.[8]

TensorFlow

TensorFlow is Google Brain's second generation machine learning system, with a reference implementation released as open source software on November 9, 2015. While the reference implementation runs on single devices, TensorFlow can run on multiple CPUs and GPUs [9] (with optional CUDA extensions for general-purpose computing on graphics processing units). It runs on 64-bit Linux or Mac OS X desktop or server systems, as well as on mobile computing platforms, including Android and Apple's iOS. TensorFlow computations are expressed as stateful dataflow graphs. Many teams at Google have migrated from DistBelief to TensorFlow for research and production uses.[4]: p.2  This library of algorithms originated from Google's need to instruct computer systems, known as neural networks, to learn and reason similarly to how humans do, so that new applications can be derived which are able to assume roles and functions previously reserved only for capable humans; the name TensorFlow itself derives from the operations which such neural networks perform on multidimensional data arrays. These multidimensional arrays are referred to as "tensors" but this concept is not identical to the mathematical concept of tensors.[10] Its purpose is to train neural networks to detect and decipher patterns and correlations.

In June 2016, Google's Jeff Dean said there were 1500 repositories on GitHub which mentioned TensorFlow, of which only 5 were from Google.[11]

Tensor processing unit (TPU)

In May 2016 Google announced its tensor processing unit (TPU), a custom ASIC built specifically for machine learning and tailored for TensorFlow. The TPU is a programmable AI accelerator designed to provide high throughput of low-precision arithmetic (e.g., 8-bit), and oriented toward using or running models rather than training them. Google announced they had been running TPUs inside their data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning.[12]

Features

TensorFlow provides a Python API, as well as a less documented C++ API.

Applications

Among the broad spectrum of applications TensorFlow is the foundation for, it has been successfully implemented in automated image captioning software, such as DeepDream.[13] Google officially implemented RankBrain on 26 October 2015, backed by TensorFlow, RankBrain now handles a substantial number of search queries, replacing and supplementing traditional static algorithm based search results.[14]

See also

References

  1. ^ a b "Credits". TensorFlow.org. Retrieved 10 November 2015.
  2. ^ "Download and Setup".
  3. ^ a b c "TensorFlow: Open source machine learning" "It is machine learning software being used for various kinds of perceptual and language understanding tasks" — Jeffrey Dean, minute 0:47 / 2:17 from Youtube clip
  4. ^ a b c Dean, Jeff; Monga, Rajat; et al. (November 9, 2015). "TensorFlow: Large-scale machine learning on heterogeneous systems" (PDF). TensorFlow.org. Google Research. Retrieved 10 November 2015.
  5. ^ Metz, Cade (November 9, 2015). "Google Just Open Sourced TensorFlow, Its Artificial Intelligence Engine". Wired. Retrieved 10 November 2015.
  6. ^ Perez, Sarah (November 9, 2015). "Google Open-Sources The Machine Learning Tech Behind Google Photos Search, Smart Reply And More". TechCrunch. Retrieved 11 November 2015.
  7. ^ Oremus, Will (November 11, 2015). "What Is TensorFlow, and Why Is Google So Excited About It?". Slate. Retrieved 11 November 2015.
  8. ^ Ward-Bailey, Jeff (November 25, 2015). "Google chairman: We're making 'real progress' on artificial intelligence". CSMonitor. Retrieved 25 November 2015.
  9. ^ Metz, Cade (November 10, 2015). "TensorFlow, Google's Open Source AI , Points to a Fast-Changing Hardware World". Wired. Retrieved 11 November 2015.
  10. ^ "Tensor". PlanetMath. Retrieved 14 February 2016.
  11. ^ Machine Learning: Google I/O 2016 Minute 07:30/44:44 accessdate=2016-06-05
  12. ^ Jouppi, Norm. "Google supercharges machine learning tasks with TPU custom chip". Google Cloud Platform Blog. Retrieved 19 May 2016.
  13. ^ Byrne, Michael (November 11, 2015). "Google Offers Up Its Entire Machine Learning Library as Open-Source Software". Vice. Retrieved 11 November 2015.
  14. ^ Woollaston, Victoria (November 25, 2015). "Google releases TensorFlow – Search giant makes its artificial intelligence software available to the public". DailyMail. Retrieved 25 November 2015.
Retrieved from "https://en.wikipedia.org/w/index.php?title=TensorFlow&oldid=752825528"