The TensorFlow implementation of FID can be found at the following link: [https://www.tensorflow.org/api_docs/python/tf/contrib/gan/eval/frechet_classifier_distance](https://www.tensorflow.org/api_docs/python/tf/contrib/gan/eval/frechet_classifier_distance)
There are more scoring algorithms available that have been recently proposed by researchers in academia and industry. We won't be covering all of these here. Before reading any further, take a look at another scoring algorithm called the Mode Score, information about which can be found at the following link: [https://arxiv.org/pdf/1612.02136.pdf](https://arxiv.org/pdf/1612.02136.pdf).
For a list of all the GANs in existence, refer to *The GAN Zoo*, an article by Avinash Hindupur available at [https://github.com/hindupuravinash/the-gan-zoo](https://github.com/hindupuravinash/the-gan-zoo).
The authors of the paper titled *Deep expectation of real and apparent age from a single image without facial landmarks,* which is available at[https://www.vision.ee.ethz.ch/en/publications/papers/articles/eth_biwi_01299.pdf](https://www.vision.ee.ethz.ch/en/publications/papers/articles/eth_biwi_01299.pdf), have scraped these images from Wikipedia and made them available for academic purposes. If you intend to use the dataset for commercial purposes, contact the authors at `rrothe@vision.ee.ethz.ch`.
We have also created a pandas `DataFrame` because it's easier to create plots using the seaborn library ([https://seaborn.pydata.org](https://seaborn.pydata.org)). In the book, the code for the plots (using Matplotlib or seaborn) is normally omitted, but it's always present in the repository.
As we have to stop with a specific TensorFlow version at certain points to get the book published, we'll continue to test run all the code in the book with every new major TensorFlow release, and update the code and test results accordingly on the book's source code repository at [http://github.com/jeffxtang/mobiletfbook](http://github.com/jeffxtang/mobiletfbook). If you have any questions about the code or the book, you may also post an issue directly on the repository.
另一个问题是 TensorFlow Mobile 和 TensorFlow Lite 之间的选择。 该书在大多数章节(1 至 10)中介绍了 TensorFlow Mobile。 TensorFlow Lite 可能是在移动设备上运行 TensorFlow 的未来,它在 Google I/O 2018 上仍处于开发人员预览版中,这就是 Google 希望您“使用 TensorFlow Mobile 覆盖生产案例”的原因。 即使在 TensorFlow Lite 正式发布后,根据 Google 的说法,“ TensorFlow Mobile 不会很快消失”-实际上,在本书出版之前我们测试了最新的 TensorFlow 1.8.0 版本,我们发现 使用 TensorFlow Mobile 变得更加简单
If you're a little confused about the buzz words machine learning, deep learning, machine intelligence, and artificial intelligence (AI), here's a quick summary: machine intelligence and AI are really just the same thing; machine learning is a field, also the most popular one, of AI; deep learning is one special type of machine learning, and is also the modern and most effective approach to solving complicated problems such as computer vision, speech recognition and synthesis, and natural language processing. So in this book, when we say AI, we primarily mean deep learning, the savior that took AI from the long winter to the summer. For more information about the AI winter and deep learning, you can check out [https://en.wikipedia.org/wiki/AI_winter](https://en.wikipedia.org/wiki/AI_winter) and [http://www.deeplearningbook.org](http://www.deeplearningbook.org).
如果您对机器学习,深度学习,机器智能和人工智能(AI)的流行语有些困惑,这里有个简短的摘要:机器智能和 AI 确实是同一回事。 机器学习是 AI 的一个领域,也是最受欢迎的领域; 深度学习是机器学习的一种特殊类型,也是解决诸如计算机视觉,语音识别和合成以及自然语言处理之类的复杂问题的现代且最有效的方法。 因此,在本书中,当我们说 AI 时,我们主要是指深度学习,这是将AI从漫长的冬天带到夏天的救星。 有关 AI 冬季和深度学习的更多信息,您可以查看[这里](https://en.wikipedia.org/wiki/AI_winter)和[这里](http://www.deeplearningbook.org)。
TensorFlow 可以安装在 MacOS,Ubuntu 或 Windows 上。 我们将介绍在 MacOS X El Capitan(10.11.6),macOS Sierra(10.12.6)和 Ubuntu 16.04 上从源代码安装 TensorFlow 1.4 的步骤。 如果您使用其他操作系统或版本,则可以参考 [TensorFlow 安装文档](https://www.tensorflow.org/install)以获取更多信息。 当您阅读本书时,可能会出现更新的 TensorFlow 版本。 尽管您仍然应该能够使用较新版本运行本书中的代码,但这并不能保证,因此我们在 Mac 和 Ubuntu 上使用 TensorFlow 1.4 发行源代码来设置 TensorFlow; 这样,您可以轻松地测试运行并与书中的应用程序一起玩。
Since we wrote the paragraph above in December 2017, there have been four new official releases of TensorFlow (1.5, 1.6, 1.7, and 1.8), which you can download at [https://github.com/tensorflow/tensorflow/releases](https://github.com/tensorflow/tensorflow/releases) or from the TensorFlow source code repo ([https://github.com/tensorflow/tensorflow](https://github.com/tensorflow/tensorflow)), and a new version of Xcode (9.3) as of May 2018\. Newer versions of TensorFlow, such as 1.8, by default support newer versions of NVIDIA CUDA and cuDNN (see the section *Setting up TensorFlow on GPU-powered Ubuntu* for detail), and you'd better follow the official TensorFlow documentation to install the latest TensorFlow version with GPU support. In this and the following chapters, we may refer to a specific TensorFlow version as an example, but will keep all iOS, Android, and Python code tested and, if needed, updated for the latest TensorFlow, Xcode, and Android Studio versions in the book's source code repo at [https://github.com/jeffxtang/mobiletfbook](https://github.com/jeffxtang/mobiletfbook).
An alternative to setting up your own GPU-powered Ubuntu with TensorFlow is to use TensorFlow in a GPU-enabled cloud service such as Google Cloud Platform's Cloud ML Engine ([https://cloud.google.com/ml-engine/docs/using-gpus](https://cloud.google.com/ml-engine/docs/using-gpus)). There are pros and cons of each option. Cloud services are generally time-based billing. If your goal is to train or retrain models to be deployed on mobile devices, meaning the models are not super complicated, and if you plan to do machine learning training for a long time, it'd be more cost effective and satisfying to have your own GPU.