简单粗暴 TensorFlow 2
follow
#
推荐序
follow
zh_hans/foreword.html
前言
follow
zh_hans/preface.html
TensorFlow概述
follow
zh_hans/introduction.html
TensorFlow安装与环境配置
follow
zh_hans/basic/installation.html
TensorFlow基础
follow
zh_hans/basic/basic.html
TensorFlow 模型建立与训练
follow
zh_hans/basic/models.html
TensorFlow常用模块
follow
zh_hans/basic/tools.html
TensorFlow模型导出
follow
zh_hans/deployment/export.html
TensorFlow Serving
follow
zh_hans/deployment/serving.html
TensorFlow Lite(Jinpeng)
follow
zh_hans/deployment/lite.html
TensorFlow in JavaScript(Huan)
follow
zh_hans/deployment/javascript.html
TensorFlow分布式训练
follow
zh_hans/appendix/distributed.html
使用TPU训练TensorFlow模型(Huan)
follow
zh_hans/appendix/tpu.html
TensorFlow Hub 模型复用(Jinpeng)
follow
zh_hans/appendix/tfhub.html
TensorFlow Datasets 数据集载入
follow
zh_hans/appendix/tfds.html
Swift for TensorFlow (S4TF) (Huan)
follow
zh_hans/appendix/swift.html
TensorFlow Quantum: 混合量子-经典机器学习 *
follow
zh_hans/appendix/quantum.html
强化学习简介
follow
zh_hans/appendix/rl.html
使用Docker部署TensorFlow环境
follow
zh_hans/appendix/docker.html
在云端使用TensorFlow
follow
zh_hans/appendix/cloud.html
部署自己的交互式Python开发环境JupyterLab
follow
zh_hans/appendix/jupyterlab.html
参考资料与推荐阅读
follow
zh_hans/appendix/recommended_books.html
术语中英对照表
follow
zh_hans/appendix/terms.html
前言
follow
zh_hant/preface.html
TensorFlow概述
follow
zh_hant/introduction.html
TensorFlow 安裝與環境配置
follow
zh_hant/basic/installation.html
TensorFlow 基礎
follow
zh_hant/basic/basic.html
TensorFlow 模型建立與訓練
follow
zh_hant/basic/models.html
TensorFlow常用模組
follow
zh_hant/basic/tools.html
TensorFlow模型匯出
follow
zh_hant/deployment/export.html
TensorFlow Serving
follow
zh_hant/deployment/serving.html
TensorFlow Lite(Jinpeng)
follow
zh_hant/deployment/lite.html
TensorFlow in JavaScript(Huan)
follow
zh_hant/deployment/javascript.html
TensorFlow分布式訓練
follow
zh_hant/appendix/distributed.html
使用TPU訓練TensorFlow模型(Huan)
follow
zh_hant/appendix/tpu.html
TensorFlow Hub 模型複用(Jinpeng)
follow
zh_hant/appendix/tfhub.html
TensorFlow Datasets 資料集載入
follow
zh_hant/appendix/tfds.html
Swift for TensorFlow (S4TF) (Huan)
follow
zh_hant/appendix/swift.html
TensorFlow Quantum: 混合量子-經典機器學習 *
follow
zh_hant/appendix/quantum.html
強化學習簡介
follow
zh_hant/appendix/rl.html
使用Docker部署TensorFlow環境
follow
zh_hant/appendix/docker.html
在雲端使用TensorFlow
follow
zh_hant/appendix/cloud.html
部署自己的互動式 Python 開發環境 JupyterLab
follow
zh_hant/appendix/jupyterlab.html
參考資料與推薦閱讀
follow
zh_hant/appendix/recommended_books.html
專有名詞中英對照表
follow
zh_hant/appendix/terms.html
Preface
follow
en/preface.html
TensorFlow Overview
follow
en/introduction.html
Installation and Environment Configuration
follow
en/basic/installation.html
TensorFlow Basic
follow
en/basic/basic.html
Model Construction and Training
follow
en/basic/models.html
Common Modules in TensorFlow
follow
en/basic/tools.html
TensorFlow Model Export
follow
en/deployment/export.html
TensorFlow Serving
follow
en/deployment/serving.html
Distributed training with TensorFlow
follow
en/appendix/distributed.html
TensorFlow Datasets: Ready-to-use Datasets
follow
en/appendix/tfds.html
TensorFlow Quantum: Hybrid Quantum-classical Machine Learning *
follow
en/appendix/quantum.html
简单粗暴 TensorFlow 2
follow
#
查看页面源码
nofollow
_sources/index.rst.txt
¶
follow
#tensorflow-2-a-concise-handbook-of-tensorflow-2
English Version
follow
/en
follow
_images/snow_leopard.jpg
《简明的 TensorFlow 2》
follow
https://item.jd.com/12980534.html
这里
follow
https://github.com/snowkylin/tensorflow-handbook/tree/master/source/_static/code
连载文章目录
follow
https://mp.weixin.qq.com/mp/appmsgalbum?action=getalbum&__biz=MzU1OTMyNDcxMQ==&scene=23&album_id=1338132220393111552#wechat_redirect
Google Summer of Code 2019
follow
https://summerofcode.withgoogle.com/archive/2019/projects/5460192307707904/
谷歌开源贡献奖(Google Open Source Peer Bonus)
follow
https://opensource.googleblog.com/2020/10/announcing-latest-google-open-source.html
https://github.com/snowkylin/tensorflow-handbook
follow
https://github.com/snowkylin/tensorflow-handbook
https://discuss.tf.wiki
follow
https://discuss.tf.wiki
follow
https://discuss.tf.wiki
https://book.douban.com/subject/35217981/
follow
https://book.douban.com/subject/35217981/
京东
follow
https://item.jd.com/12980534.html
当当
follow
http://product.dangdang.com/29132630.html
天猫
follow
https://detail.tmall.com/item.htm?id=628240887768
图灵社区
follow
https://www.ituring.com.cn/book/2705
推荐序
follow
zh_hans/foreword.html
李双峰(Google TensorFlow 中国研发负责人)
follow
zh_hans/foreword.html#google-tensorflow
Soonson Kwon (Global ML Ecosystem Programs Lead in Google)
follow
zh_hans/foreword.html#soonson-kwon-global-ml-ecosystem-programs-lead-in-google
童云海(北京大学图书馆副馆长)
follow
zh_hans/foreword.html#id4
前言
follow
zh_hans/preface.html
本书的适用群体
follow
zh_hans/preface.html#id2
如何使用本书
follow
zh_hans/preface.html#id3
致谢
follow
zh_hans/preface.html#id4
TensorFlow概述
follow
zh_hans/introduction.html
学生和研究者:模型的建立与训练
follow
zh_hans/introduction.html#id1
开发者和工程师:模型的调用与部署
follow
zh_hans/introduction.html#id2
TensorFlow能帮助我们做什么?
follow
zh_hans/introduction.html#id3
TensorFlow安装与环境配置
follow
zh_hans/basic/installation.html
一般安装步骤
follow
zh_hans/basic/installation.html#id1
GPU版本TensorFlow安装指南
follow
zh_hans/basic/installation.html#gputensorflow
GPU硬件的准备
follow
zh_hans/basic/installation.html#gpu
NVIDIA驱动程序的安装
follow
zh_hans/basic/installation.html#id5
CUDA Toolkit和cuDNN的安装
follow
zh_hans/basic/installation.html#cuda-toolkitcudnn
第一个程序
follow
zh_hans/basic/installation.html#id12
IDE设置
follow
zh_hans/basic/installation.html#ide
TensorFlow所需的硬件配置 *
follow
zh_hans/basic/installation.html#id16
TensorFlow基础
follow
zh_hans/basic/basic.html
TensorFlow 1+1
follow
zh_hans/basic/basic.html#tensorflow-1-1
自动求导机制
follow
zh_hans/basic/basic.html#zh-hans-automatic-derivation
基础示例:线性回归
follow
zh_hans/basic/basic.html#zh-hans-linear-regression
NumPy下的线性回归
follow
zh_hans/basic/basic.html#id13
TensorFlow下的线性回归
follow
zh_hans/basic/basic.html#zh-hans-optimizer
TensorFlow 模型建立与训练
follow
zh_hans/basic/models.html
模型(Model)与层(Layer)
follow
zh_hans/basic/models.html#model-layer
基础示例:多层感知机(MLP)
follow
zh_hans/basic/models.html#mlp
数据获取及预处理: tf.keras.datasets
follow
zh_hans/basic/models.html#tf-keras-datasets
模型的构建: tf.keras.Model 和 tf.keras.layers
follow
zh_hans/basic/models.html#tf-keras-model-tf-keras-layers
模型的训练: tf.keras.losses 和 tf.keras.optimizer
follow
zh_hans/basic/models.html#tf-keras-losses-tf-keras-optimizer
模型的评估: tf.keras.metrics
follow
zh_hans/basic/models.html#tf-keras-metrics
卷积神经网络(CNN)
follow
zh_hans/basic/models.html#cnn
使用Keras实现卷积神经网络
follow
zh_hans/basic/models.html#keras
使用Keras中预定义的经典卷积神经网络结构
follow
zh_hans/basic/models.html#id13
循环神经网络(RNN)
follow
zh_hans/basic/models.html#rnn
深度强化学习(DRL)
follow
zh_hans/basic/models.html#drl
Keras Pipeline *
follow
zh_hans/basic/models.html#keras-pipeline
Keras Sequential/Functional API 模式建立模型
follow
zh_hans/basic/models.html#keras-sequential-functional-api
使用 Keras Model 的 compile 、 fit 和 evaluate 方法训练和评估模型
follow
zh_hans/basic/models.html#keras-model-compile-fit-evaluate
自定义层、损失函数和评估指标 *
follow
zh_hans/basic/models.html#id24
自定义层
follow
zh_hans/basic/models.html#zh-hans-custom-layer
自定义损失函数和评估指标
follow
zh_hans/basic/models.html#id26
TensorFlow常用模块
follow
zh_hans/basic/tools.html
tf.train.Checkpoint :变量的保存与恢复
follow
zh_hans/basic/tools.html#tf-train-checkpoint
TensorBoard:训练过程可视化
follow
zh_hans/basic/tools.html#tensorboard
实时查看参数变化情况
follow
zh_hans/basic/tools.html#id1
查看Graph和Profile信息
follow
zh_hans/basic/tools.html#graphprofile
实例:查看多层感知机模型的训练情况
follow
zh_hans/basic/tools.html#id3
tf.data :数据集的构建与预处理
follow
zh_hans/basic/tools.html#tf-data
数据集对象的建立
follow
zh_hans/basic/tools.html#id4
数据集对象的预处理
follow
zh_hans/basic/tools.html#id5
使用 tf.data 的并行化策略提高训练流程效率
follow
zh_hans/basic/tools.html#zh-hans-prefetch
数据集元素的获取与使用
follow
zh_hans/basic/tools.html#id11
实例:cats_vs_dogs图像分类
follow
zh_hans/basic/tools.html#cats-vs-dogs
TFRecord :TensorFlow数据集存储格式
follow
zh_hans/basic/tools.html#tfrecord-tensorflow
将数据集存储为 TFRecord 文件
follow
zh_hans/basic/tools.html#tfrecord
读取 TFRecord 文件
follow
zh_hans/basic/tools.html#id14
tf.function :图执行模式 *
follow
zh_hans/basic/tools.html#tf-function
tf.function 基础使用方法
follow
zh_hans/basic/tools.html#id15
tf.function 内在机制
follow
zh_hans/basic/tools.html#id17
AutoGraph:将Python控制流转换为TensorFlow计算图
follow
zh_hans/basic/tools.html#autograph-pythontensorflow
使用传统的 tf.Session
follow
zh_hans/basic/tools.html#tf-session
tf.TensorArray :TensorFlow 动态数组 *
follow
zh_hans/basic/tools.html#tf-tensorarray-tensorflow
tf.config:GPU的使用与分配 *
follow
zh_hans/basic/tools.html#tf-config-gpu
指定当前程序使用的GPU
follow
zh_hans/basic/tools.html#gpu
设置显存使用策略
follow
zh_hans/basic/tools.html#id19
单GPU模拟多GPU环境
follow
zh_hans/basic/tools.html#gpugpu
TensorFlow模型导出
follow
zh_hans/deployment/export.html
使用SavedModel完整导出模型
follow
zh_hans/deployment/export.html#savedmodel
Keras 自有的模型导出格式(Jinpeng)
follow
zh_hans/deployment/export.html#keras-jinpeng
TensorFlow Serving
follow
zh_hans/deployment/serving.html
TensorFlow Serving安装
follow
zh_hans/deployment/serving.html#id1
TensorFlow Serving模型部署
follow
zh_hans/deployment/serving.html#id2
Keras Sequential模式模型的部署
follow
zh_hans/deployment/serving.html#keras-sequential
自定义Keras模型的部署
follow
zh_hans/deployment/serving.html#keras
在客户端调用以TensorFlow Serving部署的模型
follow
zh_hans/deployment/serving.html#zh-hans-call-serving-api
Python客户端示例
follow
zh_hans/deployment/serving.html#python
Node.js客户端示例(Ziyang)
follow
zh_hans/deployment/serving.html#node-js-ziyang
TensorFlow Lite(Jinpeng)
follow
zh_hans/deployment/lite.html
模型转换
follow
zh_hans/deployment/lite.html#id1
Android部署
follow
zh_hans/deployment/lite.html#android
Quantization模型转换
follow
zh_hans/deployment/lite.html#quantization
总结
follow
zh_hans/deployment/lite.html#id3
TensorFlow in JavaScript(Huan)
follow
zh_hans/deployment/javascript.html
TensorFlow.js 简介
follow
zh_hans/deployment/javascript.html#tensorflow-js
浏览器中使用 TensorFlow.js 的优势
follow
zh_hans/deployment/javascript.html#id1
TensorFlow.js 环境配置
follow
zh_hans/deployment/javascript.html#id2
在浏览器中使用 TensorFlow.js
follow
zh_hans/deployment/javascript.html#id3
在 Node.js 中使用 TensorFlow.js
follow
zh_hans/deployment/javascript.html#node-js-tensorflow-js
在微信小程序中使用 TensorFlow.js
follow
zh_hans/deployment/javascript.html#id4
TensorFlow.js 模型部署
follow
zh_hans/deployment/javascript.html#id7
在浏览器中加载 Python 模型
follow
zh_hans/deployment/javascript.html#python
在 Node.js 中执行原生 SavedModel 模型
follow
zh_hans/deployment/javascript.html#node-js-savedmodel
使用 TensorFlow.js 模型库
follow
zh_hans/deployment/javascript.html#id8
在浏览器中使用 MobileNet 进行摄像头物体识别
follow
zh_hans/deployment/javascript.html#mobilenet
TensorFlow.js 模型训练 *
follow
zh_hans/deployment/javascript.html#id9
TensorFlow.js 性能对比
follow
zh_hans/deployment/javascript.html#id13
TensorFlow分布式训练
follow
zh_hans/appendix/distributed.html
单机多卡训练: MirroredStrategy
follow
zh_hans/appendix/distributed.html#mirroredstrategy
多机训练: MultiWorkerMirroredStrategy
follow
zh_hans/appendix/distributed.html#multiworkermirroredstrategy
使用TPU训练TensorFlow模型(Huan)
follow
zh_hans/appendix/tpu.html
TPU 简介
follow
zh_hans/appendix/tpu.html#tpu
什么是 TPU
follow
zh_hans/appendix/tpu.html#id1
为什么使用 TPU
follow
zh_hans/appendix/tpu.html#id2
TPU 性能
follow
zh_hans/appendix/tpu.html#id3
TPU 环境配置
follow
zh_hans/appendix/tpu.html#id4
免费 TPU:Google Colab
follow
zh_hans/appendix/tpu.html#tpu-google-colab
Cloud TPU
follow
zh_hans/appendix/tpu.html#cloud-tpu
TPU 基础使用
follow
zh_hans/appendix/tpu.html#id5
TensorFlow Hub 模型复用(Jinpeng)
follow
zh_hans/appendix/tfhub.html
TF Hub 网站
follow
zh_hans/appendix/tfhub.html#tf-hub
TF Hub 安装
follow
zh_hans/appendix/tfhub.html#id1
TF Hub 模型使用样例
follow
zh_hans/appendix/tfhub.html#id2
TF Hub 模型retrain样例
follow
zh_hans/appendix/tfhub.html#tf-hub-retrain
TensorFlow Datasets 数据集载入
follow
zh_hans/appendix/tfds.html
Swift for TensorFlow (S4TF) (Huan)
follow
zh_hans/appendix/swift.html
S4TF 简介
follow
zh_hans/appendix/swift.html#s4tf
为什么要使用 Swift 进行 TensorFlow 开发
follow
zh_hans/appendix/swift.html#swift-tensorflow
S4TF 环境配置
follow
zh_hans/appendix/swift.html#id1
本地安装 Swift for TensorFlow
follow
zh_hans/appendix/swift.html#id2
在 Colaboratory 中快速体验 Swift for TensorFlow
follow
zh_hans/appendix/swift.html#colaboratory-swift-for-tensorflow
在 Docker 中快速体验 Swift for TensorFlow
follow
zh_hans/appendix/swift.html#docker-swift-for-tensorflow
S4TF 基础使用
follow
zh_hans/appendix/swift.html#id5
在 Swift 中使用标准的 TensorFlow API
follow
zh_hans/appendix/swift.html#swift-tensorflow-api
在 Swift 中直接加载 Python 语言库
follow
zh_hans/appendix/swift.html#swift-python
语言原生支持自动微分
follow
zh_hans/appendix/swift.html#id6
MNIST数字分类
follow
zh_hans/appendix/swift.html#mnist
TensorFlow Quantum: 混合量子-经典机器学习 *
follow
zh_hans/appendix/quantum.html
量子计算基本概念
follow
zh_hans/appendix/quantum.html#id2
量子比特
follow
zh_hans/appendix/quantum.html#id5
量子逻辑门
follow
zh_hans/appendix/quantum.html#id8
量子线路
follow
zh_hans/appendix/quantum.html#id11
多比特的量子线路和量子纠缠 *
follow
zh_hans/appendix/quantum.html#id12
实例:使用Cirq建立简单的量子线路
follow
zh_hans/appendix/quantum.html#cirq
混合量子-经典机器学习
follow
zh_hans/appendix/quantum.html#id14
量子数据集与带参数的量子门
follow
zh_hans/appendix/quantum.html#id15
参数化的量子线路(PQC)
follow
zh_hans/appendix/quantum.html#pqc
将参数化的量子线路嵌入机器学习模型
follow
zh_hans/appendix/quantum.html#id16
实例:对量子数据集进行二分类
follow
zh_hans/appendix/quantum.html#id17
强化学习简介
follow
zh_hans/appendix/rl.html
从动态规划说起
follow
zh_hans/appendix/rl.html#id2
加入随机性和概率的动态规划
follow
zh_hans/appendix/rl.html#id5
环境信息无法直接获得的情况
follow
zh_hans/appendix/rl.html#id7
从直接算法到迭代算法
follow
zh_hans/appendix/rl.html#id9
q值的渐进性更新
follow
zh_hans/appendix/rl.html#q
探索策略
follow
zh_hans/appendix/rl.html#id12
大规模问题的求解
follow
zh_hans/appendix/rl.html#id13
总结
follow
zh_hans/appendix/rl.html#id14
使用Docker部署TensorFlow环境
follow
zh_hans/appendix/docker.html
在云端使用TensorFlow
follow
zh_hans/appendix/cloud.html
在Colab中使用TensorFlow
follow
zh_hans/appendix/cloud.html#colabtensorflow
在Google Cloud Platform(GCP)中使用TensorFlow
follow
zh_hans/appendix/cloud.html#google-cloud-platform-gcp-tensorflow
在Compute Engine建立带GPU的实例并部署TensorFlow
follow
zh_hans/appendix/cloud.html#compute-enginegputensorflow
使用AI Platform中的Notebook建立带GPU的在线JupyterLab环境
follow
zh_hans/appendix/cloud.html#ai-platformnotebookgpujupyterlab
在阿里云上使用 GPU 实例运行 Tensorflow(Ziyang)
follow
zh_hans/appendix/cloud.html#gpu-tensorflow-ziyang
部署自己的交互式Python开发环境JupyterLab
follow
zh_hans/appendix/jupyterlab.html
参考资料与推荐阅读
follow
zh_hans/appendix/recommended_books.html
术语中英对照表
follow
zh_hans/appendix/terms.html
前言
follow
zh_hant/preface.html
本書的適用羣體
follow
zh_hant/preface.html#id2
如何使用本書
follow
zh_hant/preface.html#id3
致謝
follow
zh_hant/preface.html#id4
TensorFlow概述
follow
zh_hant/introduction.html
學生和研究者:模型的建立與訓練
follow
zh_hant/introduction.html#id1
開發者和工程師:模型的呼叫與部署
follow
zh_hant/introduction.html#id2
TensorFlow 能幫助我們做什麼?
follow
zh_hant/introduction.html#id3
TensorFlow 安裝與環境配置
follow
zh_hant/basic/installation.html
一般安裝步驟
follow
zh_hant/basic/installation.html#id1
GPU 版本 TensorFlow 安裝指南
follow
zh_hant/basic/installation.html#gpu-tensorflow
GPU 硬體的準備
follow
zh_hant/basic/installation.html#gpu
NVIDIA 驅動程式的安裝
follow
zh_hant/basic/installation.html#id6
CUDA Toolkit 和 cuDNN 的安裝
follow
zh_hant/basic/installation.html#cuda-toolkit-cudnn
第一個程式
follow
zh_hant/basic/installation.html#id13
IDE 設置
follow
zh_hant/basic/installation.html#ide
TensorFlow 所需的硬體配置 *
follow
zh_hant/basic/installation.html#id17
TensorFlow 基礎
follow
zh_hant/basic/basic.html
TensorFlow 1+1
follow
zh_hant/basic/basic.html#tensorflow-1-1
自動推導機制
follow
zh_hant/basic/basic.html#automatic-derivation
基礎範例:線性回歸
follow
zh_hant/basic/basic.html#linear-regression
NumPy 下的線性回歸
follow
zh_hant/basic/basic.html#id16
TensorFlow下的線性回歸
follow
zh_hant/basic/basic.html#optimizer
TensorFlow 模型建立與訓練
follow
zh_hant/basic/models.html
模型(Model)與層(Layer)
follow
zh_hant/basic/models.html#model-layer
基礎範例:多層感知器(MLP)
follow
zh_hant/basic/models.html#mlp
資料獲取及預處理: tf.keras.datasets
follow
zh_hant/basic/models.html#tf-keras-datasets
模型的建構: tf.keras.Model 和 tf.keras.layers
follow
zh_hant/basic/models.html#tf-keras-model-tf-keras-layers
模型的訓練: tf.keras.losses 和 tf.keras.optimizer
follow
zh_hant/basic/models.html#tf-keras-losses-tf-keras-optimizer
模型的評估: tf.keras.metrics
follow
zh_hant/basic/models.html#tf-keras-metrics
卷積神經網路(CNN)
follow
zh_hant/basic/models.html#cnn
使用Keras實現卷積神經網路
follow
zh_hant/basic/models.html#keras
使用Keras中預定義的典型卷積神經網路結構
follow
zh_hant/basic/models.html#id14
循環神經網路(RNN)
follow
zh_hant/basic/models.html#rnn
深度強化學習(DRL)
follow
zh_hant/basic/models.html#drl
Keras Pipeline *
follow
zh_hant/basic/models.html#keras-pipeline
Keras Sequential/Functional API 模式建立模型
follow
zh_hant/basic/models.html#keras-sequential-functional-api
使用 Keras Model 的 compile 、 fit 和 evaluate 方法訓練和評估模型
follow
zh_hant/basic/models.html#keras-model-compile-fit-evaluate
自定義層、損失函數和評量指標 *
follow
zh_hant/basic/models.html#id23
自定義層
follow
zh_hant/basic/models.html#custom-layer
自定義損失函數和評量指標
follow
zh_hant/basic/models.html#id25
TensorFlow常用模組
follow
zh_hant/basic/tools.html
tf.train.Checkpoint :變數的保存與還原
follow
zh_hant/basic/tools.html#tf-train-checkpoint
TensorBoard:訓練過程可視化
follow
zh_hant/basic/tools.html#tensorboard
實時查看參數變化情況
follow
zh_hant/basic/tools.html#id1
查看Graph和Profile信息
follow
zh_hant/basic/tools.html#graphprofile
實例:查看多層感知器模型的訓練情況
follow
zh_hant/basic/tools.html#id3
tf.data :資料集的建立與預處理
follow
zh_hant/basic/tools.html#tf-data
資料集對象的建立
follow
zh_hant/basic/tools.html#id4
資料集對象的預處理
follow
zh_hant/basic/tools.html#id5
使用 tf.data 的平行化策略提高訓練流程效率
follow
zh_hant/basic/tools.html#prefetch
資料集元素的獲取與使用
follow
zh_hant/basic/tools.html#id11
實例:cats_vs_dogs圖片分類
follow
zh_hant/basic/tools.html#cats-vs-dogs
TFRecord :TensorFlow資料集存儲格式
follow
zh_hant/basic/tools.html#tfrecord-tensorflow
將資料集存儲為 TFRecord 文件
follow
zh_hant/basic/tools.html#id14
讀取 TFRecord 文件
follow
zh_hant/basic/tools.html#id16
tf.function :圖執行模式 *
follow
zh_hant/basic/tools.html#tf-function
tf.function 基礎使用方法
follow
zh_hant/basic/tools.html#id17
tf.function 內在機制
follow
zh_hant/basic/tools.html#id19
AutoGraph:將Python控制流轉換為TensorFlow計算圖
follow
zh_hant/basic/tools.html#autograph-pythontensorflow
使用傳統的 tf.Session
follow
zh_hant/basic/tools.html#tf-session
tf.TensorArray :TensorFlow 動態陣列 *
follow
zh_hant/basic/tools.html#tf-tensorarray-tensorflow
tf.config:GPU的使用與分配 *
follow
zh_hant/basic/tools.html#tf-config-gpu
指定當前程式使用的GPU
follow
zh_hant/basic/tools.html#gpu
設置顯示卡記憶體儲存空間使用策略
follow
zh_hant/basic/tools.html#id21
單GPU模擬多GPU環境
follow
zh_hant/basic/tools.html#gpugpu
TensorFlow模型匯出
follow
zh_hant/deployment/export.html
使用SavedModel完整匯出模型
follow
zh_hant/deployment/export.html#savedmodel
Keras Sequential save方法(Jinpeng)
follow
zh_hant/deployment/export.html#keras-sequential-save-jinpeng
TensorFlow Serving
follow
zh_hant/deployment/serving.html
TensorFlow Serving 安裝
follow
zh_hant/deployment/serving.html#id1
TensorFlow Serving 模型部署
follow
zh_hant/deployment/serving.html#id2
Keras Sequential 模式模型的部署
follow
zh_hant/deployment/serving.html#keras-sequential
自定義 Keras 模型的部署
follow
zh_hant/deployment/serving.html#keras
在客戶端呼叫以 TensorFlow Serving 部署的模型
follow
zh_hant/deployment/serving.html#call-serving-api
Python 客戶端範例
follow
zh_hant/deployment/serving.html#python
Node.js客戶端範例(Ziyang)
follow
zh_hant/deployment/serving.html#node-js-ziyang
TensorFlow Lite(Jinpeng)
follow
zh_hant/deployment/lite.html
模型轉換
follow
zh_hant/deployment/lite.html#id1
Android部署
follow
zh_hant/deployment/lite.html#android
Quantization 模型轉換
follow
zh_hant/deployment/lite.html#quantization
總結
follow
zh_hant/deployment/lite.html#id3
TensorFlow in JavaScript(Huan)
follow
zh_hant/deployment/javascript.html
TensorFlow.js 簡介
follow
zh_hant/deployment/javascript.html#tensorflow-js
瀏覽器中使用 TensorFlow.js 的優勢
follow
zh_hant/deployment/javascript.html#id1
TensorFlow.js 環境配置
follow
zh_hant/deployment/javascript.html#id2
在瀏覽器中使用 TensorFlow.js
follow
zh_hant/deployment/javascript.html#id3
在 Node.js 中使用 TensorFlow.js
follow
zh_hant/deployment/javascript.html#node-js-tensorflow-js
在微信小程式中使用 TensorFlow.js
follow
zh_hant/deployment/javascript.html#id4
TensorFlow.js 模型部署
follow
zh_hant/deployment/javascript.html#id7
在瀏覽器中加載 Python 模型
follow
zh_hant/deployment/javascript.html#python
在 Node.js 中執行原生 SavedModel 模型
follow
zh_hant/deployment/javascript.html#node-js-savedmodel
使用 TensorFlow.js 模型資料庫
follow
zh_hant/deployment/javascript.html#id8
在瀏覽器中使用 MobileNet 進行攝像頭物體辨識
follow
zh_hant/deployment/javascript.html#mobilenet
TensorFlow.js 模型訓練 *
follow
zh_hant/deployment/javascript.html#id11
TensorFlow.js 性能對比
follow
zh_hant/deployment/javascript.html#id15
TensorFlow分布式訓練
follow
zh_hant/appendix/distributed.html
單機多卡訓練: MirroredStrategy
follow
zh_hant/appendix/distributed.html#mirroredstrategy
多機訓練: MultiWorkerMirroredStrategy
follow
zh_hant/appendix/distributed.html#multiworkermirroredstrategy
使用TPU訓練TensorFlow模型(Huan)
follow
zh_hant/appendix/tpu.html
TPU 簡介
follow
zh_hant/appendix/tpu.html#tpu
什麼是 TPU
follow
zh_hant/appendix/tpu.html#id1
爲什麼使用 TPU
follow
zh_hant/appendix/tpu.html#id2
TPU 性能
follow
zh_hant/appendix/tpu.html#id3
TPU 環境配置
follow
zh_hant/appendix/tpu.html#id4
免費 TPU:Google Colab
follow
zh_hant/appendix/tpu.html#tpu-google-colab
Cloud TPU
follow
zh_hant/appendix/tpu.html#cloud-tpu
TensorFlow Hub 模型複用(Jinpeng)
follow
zh_hant/appendix/tfhub.html
TF Hub 網站
follow
zh_hant/appendix/tfhub.html#tf-hub
TF Hub 安裝
follow
zh_hant/appendix/tfhub.html#id1
TF Hub 模型使用案例
follow
zh_hant/appendix/tfhub.html#id2
TF Hub 模型 Retrain 範例
follow
zh_hant/appendix/tfhub.html#tf-hub-retrain
TensorFlow Datasets 資料集載入
follow
zh_hant/appendix/tfds.html
Swift for TensorFlow (S4TF) (Huan)
follow
zh_hant/appendix/swift.html
S4TF 簡介
follow
zh_hant/appendix/swift.html#s4tf
為什麼要使用 Swift 進行 TensorFlow 開發
follow
zh_hant/appendix/swift.html#swift-tensorflow
S4TF 環境配置
follow
zh_hant/appendix/swift.html#id1
本機安裝 Swift for TensorFlow
follow
zh_hant/appendix/swift.html#id2
在 Colaboratory 中快速體驗 Swift for TensorFlow
follow
zh_hant/appendix/swift.html#colaboratory-swift-for-tensorflow
在 Docker 中快速體驗 Swift for TensorFlow
follow
zh_hant/appendix/swift.html#docker-swift-for-tensorflow
S4TF 基礎使用
follow
zh_hant/appendix/swift.html#id5
在 Swift 中使用標準的 TensorFlow API
follow
zh_hant/appendix/swift.html#swift-tensorflow-api
在 Swift 中直接載入 Python 語言函式庫
follow
zh_hant/appendix/swift.html#swift-python
語言原生支持自動微分
follow
zh_hant/appendix/swift.html#id6
MNIST數字分類
follow
zh_hant/appendix/swift.html#mnist
TensorFlow Quantum: 混合量子-經典機器學習 *
follow
zh_hant/appendix/quantum.html
量子計算基本概念
follow
zh_hant/appendix/quantum.html#id1
量子位元
follow
zh_hant/appendix/quantum.html#id2
量子邏輯閘
follow
zh_hant/appendix/quantum.html#id5
量子電路
follow
zh_hant/appendix/quantum.html#id8
實例:使用Cirq建立簡單的量子電路
follow
zh_hant/appendix/quantum.html#cirq
混合量子-經典機器學習
follow
zh_hant/appendix/quantum.html#id10
量子資料集與包含參數的量子閘
follow
zh_hant/appendix/quantum.html#id11
參數化的量子電路(PQC)
follow
zh_hant/appendix/quantum.html#pqc
將參數化的量子電路嵌入機器學習模型
follow
zh_hant/appendix/quantum.html#id12
範例:對量子資料集進行二分類
follow
zh_hant/appendix/quantum.html#id13
強化學習簡介
follow
zh_hant/appendix/rl.html
從動態規劃說起
follow
zh_hant/appendix/rl.html#id2
加入隨機性和機率的動態規劃
follow
zh_hant/appendix/rl.html#id5
環境資訊無法直接獲得的情況
follow
zh_hant/appendix/rl.html#id7
從直接演算法到疊代演算法
follow
zh_hant/appendix/rl.html#id9
q值的漸進性更新
follow
zh_hant/appendix/rl.html#q
探索策略
follow
zh_hant/appendix/rl.html#id12
大規模問題的求解
follow
zh_hant/appendix/rl.html#id13
總結
follow
zh_hant/appendix/rl.html#id14
使用Docker部署TensorFlow環境
follow
zh_hant/appendix/docker.html
在雲端使用TensorFlow
follow
zh_hant/appendix/cloud.html
在Colab中使用TensorFlow
follow
zh_hant/appendix/cloud.html#colabtensorflow
在Google Cloud Platform(GCP)中使用TensorFlow
follow
zh_hant/appendix/cloud.html#google-cloud-platform-gcp-tensorflow
在 Compute Engine 建立支援 GPU 的實例並部署 TensorFlow
follow
zh_hant/appendix/cloud.html#compute-engine-gpu-tensorflow
使用 AI Platform 中的 Notebook 建立資源GPU 的線上 JupyterLab 環境
follow
zh_hant/appendix/cloud.html#ai-platform-notebook-gpu-jupyterlab
部署自己的互動式 Python 開發環境 JupyterLab
follow
zh_hant/appendix/jupyterlab.html
參考資料與推薦閱讀
follow
zh_hant/appendix/recommended_books.html
專有名詞中英對照表
follow
zh_hant/appendix/terms.html
¶
follow
#english-version-in-progress
here
follow
https://github.com/snowkylin/tensorflow-handbook/tree/master/source/_static/code/en
https://v1.tf.wiki
follow
https://v1.tf.wiki
Google Summer of Code 2019
follow
https://summerofcode.withgoogle.com/archive/2019/projects/5460192307707904/
https://github.com/snowkylin/tensorflow-handbook
follow
https://github.com/snowkylin/tensorflow-handbook
https://discuss.tf.wiki
follow
https://discuss.tf.wiki
Preface
follow
en/preface.html
Target readers
follow
en/preface.html#target-readers
Usage
follow
en/preface.html#usage
Acknowledgement
follow
en/preface.html#acknowledgement
TensorFlow Overview
follow
en/introduction.html
For students and researchers: To build and train models
follow
en/introduction.html#for-students-and-researchers-to-build-and-train-models
For developers and engineers: To call and deploy models
follow
en/introduction.html#for-developers-and-engineers-to-call-and-deploy-models
What can TensorFlow do for us?
follow
en/introduction.html#what-can-tensorflow-do-for-us
Installation and Environment Configuration
follow
en/basic/installation.html
General steps for installation
follow
en/basic/installation.html#general-steps-for-installation
Guide for TensorFlow GPU version installation
follow
en/basic/installation.html#guide-for-tensorflow-gpu-version-installation
Preperations for GPU hardwares
follow
en/basic/installation.html#preperations-for-gpu-hardwares
Installation of NVIDIA drivers
follow
en/basic/installation.html#installation-of-nvidia-drivers
Installation of CUDA Toolkit and cuDNN
follow
en/basic/installation.html#installation-of-cuda-toolkit-and-cudnn
Your first program
follow
en/basic/installation.html#your-first-program
IDE configuration
follow
en/basic/installation.html#ide-configuration
The hardware configuration for TensorFlow *
follow
en/basic/installation.html#the-hardware-configuration-for-tensorflow
TensorFlow Basic
follow
en/basic/basic.html
TensorFlow 1+1
follow
en/basic/basic.html#tensorflow-1-1
Automatic differentiation mechanism
follow
en/basic/basic.html#automatic-differentiation-mechanism
A basic example: Linear regression
follow
en/basic/basic.html#a-basic-example-linear-regression
Linear regression under NumPy
follow
en/basic/basic.html#linear-regression-under-numpy
Linear regression under TensorFlow
follow
en/basic/basic.html#linear-regression-under-tensorflow
Model Construction and Training
follow
en/basic/models.html
Models and layers
follow
en/basic/models.html#models-and-layers
Basic example: multi-layer perceptron (MLP)
follow
en/basic/models.html#basic-example-multi-layer-perceptron-mlp
Data acquisition and pre-processing with tf.keras.datasets
follow
en/basic/models.html#data-acquisition-and-pre-processing-with-tf-keras-datasets
Model construction with tf.keras.Model and tf.keras.layers
follow
en/basic/models.html#model-construction-with-tf-keras-model-and-tf-keras-layers
Model training with tf.keras.losses and tf.keras.optimizer
follow
en/basic/models.html#model-training-with-tf-keras-losses-and-tf-keras-optimizer
Model Evaluation with tf.keras.metrics
follow
en/basic/models.html#model-evaluation-with-tf-keras-metrics
Convolutional Neural Network (CNN)
follow
en/basic/models.html#convolutional-neural-network-cnn
Implementing Convolutional Neural Networks with Keras
follow
en/basic/models.html#implementing-convolutional-neural-networks-with-keras
Using predefined classical CNN structures in Keras
follow
en/basic/models.html#using-predefined-classical-cnn-structures-in-keras
Recurrent Neural Network (RNN)
follow
en/basic/models.html#recurrent-neural-network-rnn
Deep Reinforcement Learning (DRL)
follow
en/basic/models.html#deep-reinforcement-learning-drl
Keras Pipeline *
follow
en/basic/models.html#keras-pipeline
Use Keras Sequential/Functional API to build models
follow
en/basic/models.html#use-keras-sequential-functional-api-to-build-models
Train and evaluate models using the compile, fit and evaluate methods of Keras
follow
en/basic/models.html#train-and-evaluate-models-using-the-compile-fit-and-evaluate-methods-of-keras
Custom layers, losses and metrics *
follow
en/basic/models.html#custom-layers-losses-and-metrics
Custom layers
follow
en/basic/models.html#custom-layers
Custom loss functions and metrics
follow
en/basic/models.html#custom-loss-functions-and-metrics
Common Modules in TensorFlow
follow
en/basic/tools.html
Variable saving and restore: tf.train.Checkpoint
follow
en/basic/tools.html#variable-saving-and-restore-tf-train-checkpoint
Visualization of training process: TensorBoard
follow
en/basic/tools.html#visualization-of-training-process-tensorboard
Real-time monitoring of indicator change
follow
en/basic/tools.html#real-time-monitoring-of-indicator-change
Visualize Graph and Profile Information
follow
en/basic/tools.html#visualize-graph-and-profile-information
Example: visualize the training process of MLP
follow
en/basic/tools.html#example-visualize-the-training-process-of-mlp
Dataset construction and preprocessing: tf.data
follow
en/basic/tools.html#dataset-construction-and-preprocessing-tf-data
Dataset construction
follow
en/basic/tools.html#dataset-construction
Dataset preprocessing
follow
en/basic/tools.html#dataset-preprocessing
Increase the efficiency using the parallelization strategy of tf.data
follow
en/basic/tools.html#increase-the-efficiency-using-the-parallelization-strategy-of-tf-data
Fetching elements from datasets
follow
en/basic/tools.html#fetching-elements-from-datasets
Example: cats_vs_dogs image classification
follow
en/basic/tools.html#example-cats-vs-dogs-image-classification
TFRecord: Dataset format of TensorFlow
follow
en/basic/tools.html#tfrecord-dataset-format-of-tensorflow
Convert the dataset into a TFRecord file
follow
en/basic/tools.html#convert-the-dataset-into-a-tfrecord-file
Read the TFRecord file
follow
en/basic/tools.html#read-the-tfrecord-file
Graph execution mode: @tf.function *
follow
en/basic/tools.html#graph-execution-mode-tf-function
Basic usage of tf.function
follow
en/basic/tools.html#basic-usage-of-tf-function
Internal mechanism of tf.function
follow
en/basic/tools.html#internal-mechanism-of-tf-function
AutoGraph: Converting Python control flows into TensorFlow graphs
follow
en/basic/tools.html#autograph-converting-python-control-flows-into-tensorflow-graphs
Using traditional tf.Session
follow
en/basic/tools.html#using-traditional-tf-session
TensorFlow dynamic array: tf.TensorArray *
follow
en/basic/tools.html#tensorflow-dynamic-array-tf-tensorarray
Setting and allocating GPUs: tf.config *
follow
en/basic/tools.html#setting-and-allocating-gpus-tf-config
Allocate GPUs for current program
follow
en/basic/tools.html#allocate-gpus-for-current-program
Setting GPU memory usage policy
follow
en/basic/tools.html#setting-gpu-memory-usage-policy
Simulating a multi-GPU environment with a single GPU
follow
en/basic/tools.html#simulating-a-multi-gpu-environment-with-a-single-gpu
TensorFlow Model Export
follow
en/deployment/export.html
Export models by SavedModel
follow
en/deployment/export.html#export-models-by-savedmodel
TensorFlow Serving
follow
en/deployment/serving.html
Installation of TensorFlow Serving
follow
en/deployment/serving.html#installation-of-tensorflow-serving
TensorFlow Serving models deployment
follow
en/deployment/serving.html#tensorflow-serving-models-deployment
Keras Sequential mode models deployment
follow
en/deployment/serving.html#keras-sequential-mode-models-deployment
Custom Keras models deployment
follow
en/deployment/serving.html#custom-keras-models-deployment
Calling models deployed by TensorFlow Serving on client
follow
en/deployment/serving.html#calling-models-deployed-by-tensorflow-serving-on-client
An example of Python client
follow
en/deployment/serving.html#an-example-of-python-client
An example of Node.js client
follow
en/deployment/serving.html#an-example-of-node-js-client
Distributed training with TensorFlow
follow
en/appendix/distributed.html
Training on a single machine with multiple GPUs: MirroredStrategy
follow
en/appendix/distributed.html#training-on-a-single-machine-with-multiple-gpus-mirroredstrategy
Training on multiple machines: MultiWorkerMirroredStrategy
follow
en/appendix/distributed.html#training-on-multiple-machines-multiworkermirroredstrategy
TensorFlow Datasets: Ready-to-use Datasets
follow
en/appendix/tfds.html
TensorFlow Quantum: Hybrid Quantum-classical Machine Learning *
follow
en/appendix/quantum.html
Basic concepts of quantum computing
follow
en/appendix/quantum.html#basic-concepts-of-quantum-computing
Quantum bit
follow
en/appendix/quantum.html#quantum-bit
Quantum logic gate
follow
en/appendix/quantum.html#quantum-logic-gate
Quantum circuit
follow
en/appendix/quantum.html#quantum-circuit
Example: Create a simple circuit circuit using Cirq
follow
en/appendix/quantum.html#example-create-a-simple-circuit-circuit-using-cirq
Hybrid Quantum - Classical Machine Learning
follow
en/appendix/quantum.html#hybrid-quantum-classical-machine-learning
Quantum datasets and quantum gates with parameters
follow
en/appendix/quantum.html#quantum-datasets-and-quantum-gates-with-parameters
Parametric quantum circuit (PQC)
follow
en/appendix/quantum.html#parametric-quantum-circuit-pqc
Embedding parametric quantum circuits into machine learning models
follow
en/appendix/quantum.html#embedding-parametric-quantum-circuits-into-machine-learning-models
Example: binary classification of quantum datasets
follow
en/appendix/quantum.html#example-binary-classification-of-quantum-datasets
¶
follow
#indices-and-tables
模块索引
follow
py-modindex.html
下一页
follow
zh_hans/foreword.html
Sphinx
follow
https://www.sphinx-doc.org/
theme
follow
https://github.com/rtfd/sphinx_rtd_theme
Read the Docs
follow
https://readthedocs.org
沪ICP备13038357号-18
follow
https://beian.miit.gov.cn/