🕷️ Crawler Inspector

URL Lookup

Direct Parameter Lookup

Raw Queries and Responses

1. Shard Calculation

Query:

Response:

Calculated Shard: 66 (from laksa033)

2. Crawled Status Check

Query:

curl -X POST \
  'http://laksa066.int.ahrefs:8124/' \
  -H 'Content-Type: text/plain' \
  -H 'X-ClickHouse-Database: crawler3' \
  -H 'Authorization: Basic YXBpOg==' \
  -d 'SELECT getAhrefsURLFromUnparsed(src_unparsed) AS found_url, ifNull(toUnixTimestamp(download_stamp), 0) AS crawl_time, ifNull(toUnixTimestamp(props_url_first_seen), 0) AS first_indexed_time, download_http_code AS http_code, src_unparsed AS src_unparsed, src_root_hash AS src_root_hash, history_drop_reason AS history_drop_reason, meta_title AS meta_title, meta_descriptions AS meta_descriptions, attrs_boilerpipe_text AS attrs_boilerpipe_text, attrs_markdown AS attrs_markdown, attrs_readable_markdown AS attrs_readable_markdown, meta_canonical AS meta_canonical FROM crawler3.page_info_local FINAL PREWHERE (src_root_hash, src_unparsed) IN ((getAhrefsRootHashFromUnparsed(getAhrefsUnparsedNoserviceFromURL(\'https://segmentfault.com/a/1190000019541931\')), getAhrefsUnparsedNoserviceFromURL(\'https://segmentfault.com/a/1190000019541931\'))) FORMAT JSONEachRow'

Response:

{"found_url":"https:\/\/segmentfault.com\/a\/1190000019541931","crawl_time":1775877304,"first_indexed_time":1561118646,"http_code":200,"src_unparsed":"com,segmentfault!\/a\/1190000019541931 s443","src_root_hash":"4631572693029667066","history_drop_reason":null,"meta_title":"python - TensorFlow 2.0 \/ TF2.0 入门教程实战案例 - 极客兔兔 - SegmentFault 思否","meta_descriptions":["用最白话的语言，讲解机器学习、神经网络与深度学习示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现相关链接一篇文章入门 Python机器学习笔试面试题，Github..."],"attrs_boilerpipe_text":"用最白话的语言，讲解机器学习、神经网络与深度学习\n示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现\n相关链接\n一篇文章入门 Python\n机器学习笔试面试题\n，\nGithub\nTensorFlow 2.0 中文文档\n，\nGithub\nTensorFlow 2.0 图像识别&强化学习实战\n，\nGithub\nOpenAI gym\nTensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient\nGithub - gym\/CartPole-v0-policy-gradient\n介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0\nTensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car\nGithub - gym\/MountainCar-v0-dqn\n介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏\nQ-Table用神经网络来代替。\nTensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym\nGithub - gym\/MountainCar-v0-q-learning\n介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏\n将连续的状态离散化。\nTensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game\nGithub - gym\/CartPole-v0-nn\n介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏\n使用TensorFlow 2.0\nmnist\nTensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络)\nGithub - v4_cnn\n介绍了如何搭建CNN网络，准确率达到0.99\n使用TensorFlow 2.0\nTensorFlow入门(四) - mnist手写数字识别(制作h5py训练集)\nGithub - make_data_set\n介绍了如何使用 numpy 制作 npy 格式的数据集\n介绍了如何使用 h5py 制作 HDF5 格式的数据集\nTensorFlow入门(三) - mnist手写数字识别(可视化训练)\nGithub - mnist\/v3\n介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图\nTensorFlow入门(二) - mnist手写数字识别(模型保存加载)\nGithub - mnist\/v2\n介绍了 TensorFlow 中如何保存训练好的模型\n介绍了如何从某一个模型为起点继续训练\n介绍了模型如何加载使用，传入真实的图片如何识别\nTensorFlow入门(一) - mnist手写数字识别(网络搭建)\nGithub - mnist\/v1\n这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。\n包括输入输出、独热编码与损失函数，以及正确率的验证。","attrs_markdown":"[极客兔兔](https:\/\/segmentfault.com\/blog\/geektutu)\n\n[极客兔兔](https:\/\/segmentfault.com\/blog\/geektutu)\n\n[注册登录](https:\/\/segmentfault.com\/user\/login)\n\n[关注博客](https:\/\/segmentfault.com\/a\/1190000019541931)[注册登录](https:\/\/segmentfault.com\/user\/login)\n\n[主页](https:\/\/segmentfault.com\/blog\/geektutu)\n\n[关于](https:\/\/segmentfault.com\/blog\/geektutu\/about)\n\n[RSS](https:\/\/segmentfault.com\/feeds\/blog\/geektutu)\n\n# [TensorFlow 2.0 \/ TF2.0 入门教程实战案例](https:\/\/segmentfault.com\/a\/1190000019541931)\n[![头像](https:\/\/avatar-static.segmentfault.com\/417\/773\/417773404-59a19cc559425_huge128)**极客兔兔**](https:\/\/segmentfault.com\/u\/geektutu)\n\n[2019-06-21](https:\/\/segmentfault.com\/a\/1190000019541931\/revision)\n\n阅读 2 分钟\n\n3\n\n> 用最白话的语言，讲解机器学习、神经网络与深度学习  \n> 示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现\n## 相关链接\n- [一篇文章入门 Python](https:\/\/link.segmentfault.com\/?enc=zO40sB6JYWO0CUkCVdslqQ%3D%3D.erm5mTbN70JccY0V1ZfNiOfS%2FvlzHt4vDOs%2Bbij29Ii0GlapYuMxwTgmBLcvO3zw)\n- [机器学习笔试面试题](https:\/\/link.segmentfault.com\/?enc=Rqjl6qSgJgF0v28D20EELA%3D%3D.49JXB6s%2Fcue6%2B1pUi8sf4T1HM3iRSmHarhbYdvrY%2BlpGPUg9gk%2Fl8lEWtKwP48M%2B)，[Github](https:\/\/link.segmentfault.com\/?enc=q91i1qmN60N6Poq3sDGnCg%3D%3D.UCINOkmzdCPCeThxkR%2BPCa5aoSr20ag5TokdDUN%2BVQCU1Bi6atDYbXRwZYYIqXYe)\n- [TensorFlow 2.0 中文文档](https:\/\/link.segmentfault.com\/?enc=NzldmwAnmuqJfKZbg5L0DQ%3D%3D.UPaRQ8aMA1dcYscUyiDu%2BzzluIJHBcXVRPB4FDMsAJCXJfF2T8BWX1LntvX%2FIz3P)，[Github](https:\/\/link.segmentfault.com\/?enc=h2S8L9peI%2FMVYktBI6wysQ%3D%3D.DMC%2BN28Lz7QhbqL80m72eIUw4PfUZJX0csR88olqCcOTOsvnPCV02htIG%2F%2Ft9M0G)\n- [TensorFlow 2.0 图像识别&强化学习实战](https:\/\/link.segmentfault.com\/?enc=ylYxIgxvKgmG1oTC3ynj%2BA%3D%3D.HqX%2FibuVDWOFm0Hr5L2IIt3LEgTYj0oAO2I5T8YtrZG2arQOT3%2F9KTOyDe6RVeqsBIvTM5DwoeiNd15MDFTFcw%3D%3D)，[Github](https:\/\/link.segmentfault.com\/?enc=rdFycR6KsBWOFxj1E6rrFQ%3D%3D.%2F95LbrTMnnaStIhn2JVO%2FyQpDVhQ85E2Hlhfv7mSY2KPP%2FNwLuTpSBYDGsEYh8bPjZk3zcNgBqeWjeDonHCb9g%3D%3D)\n## OpenAI gym\n- [TensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient](https:\/\/link.segmentfault.com\/?enc=1Zl6QZI7bxGrZSY%2BVr0Cvg%3D%3D.qfag0sCR9mlSSlbsuQ44GblRZBmaTTo0lJPQd%2Bctk1tEAzYRo99URx2S3Hol7t5fuhK%2BDjf9wTgR3qsruGoUPg%3D%3D)\n  - [Github - gym\/CartPole-v0-policy-gradient](https:\/\/link.segmentfault.com\/?enc=oB9emap3au%2BbKUUfaT%2F5qw%3D%3D.BWnZrCqluh2cbUZ06RPCbG3kCB5Hv03UMY8r7U5Qp9m1o%2FHas5ouVZMPj3Ygx9S%2B0Yr%2BoaVprIHl1j1YuwJOZOwogmJ8L2wqdheOdoJXFHIBK5rV%2FgI%2B57U1oFPtk28SmFO%2FJZ1VKvJ8amHdnmHMmg%3D%3D)\n  - 介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0\n- [TensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car](https:\/\/link.segmentfault.com\/?enc=63mByzrb9M%2FRkmpAqVHILg%3D%3D.6CXQC2Jmi4wbEu7YT3KK5mNOC2nUoOPe2HpVk5LqFYEYz92YikCZJTGvPCCCyJalkmYTFnJmT7P4Gy4OKp16VA%3D%3D)\n  - [Github - gym\/MountainCar-v0-dqn](https:\/\/link.segmentfault.com\/?enc=WJNlCqmmnEqjUHeQCMVhow%3D%3D.1amA5Uc21lBH754MZTgdmAnwh8qG%2Fux1Ib88U6lZBbpZNqv6Epx2JnQ5F6m7TsoDVgqsjfPmoEoT8FHyELC%2Fw2HHKOGFzE2F3rCwK%2B8RI9LKI5A%2FwD%2B%2FlpDojLJAnODf)\n  - 介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏\n  - Q-Table用神经网络来代替。\n- [TensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym](https:\/\/link.segmentfault.com\/?enc=eODVi5aVheibmzAPf7b5YA%3D%3D.mPK%2FoXCImhFAMGAtYcv3EL8JSsA0fJn%2F1kd39wpVDIreVD9W90u6QeX0a%2BBXvdaq7iK2NQ%2FtExlJHqMKcYiXcQ%3D%3D)\n  - [Github - gym\/MountainCar-v0-q-learning](https:\/\/link.segmentfault.com\/?enc=0UtAsjtjuoULsN3iJZV9qg%3D%3D.ByalkcOoAiq6NRvN%2FXOezAtfxk9Ph17q5C8Ii7Rr6%2B5uV8ojZ4%2BtF2WVFlENLGaGa2J6WH2xs6L75yvzeNQW8hOTmgwp8OhNToPcAK8%2FwVc0G0AjUnS4HuWDpkm56vbe7qhgIYMMVY631dSXt9%2BhdA%3D%3D)\n  - 介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏\n  - 将连续的状态离散化。\n- [TensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game](https:\/\/link.segmentfault.com\/?enc=MuNqKigi1bC3ZZjMqVI%2BwA%3D%3D.aTX6kHUjZXQw%2BrQBBlDblKGljtuowa5aQjCmJeGcy5cpw5JZE%2BNpnt3L7KLq390q4rd%2B9TfsxVPQsdehr3HduQ%3D%3D)\n  - [Github - gym\/CartPole-v0-nn](https:\/\/link.segmentfault.com\/?enc=SUJ7Eh4qv48fOFj5q8OsLw%3D%3D.R%2BNQFxUEMHdCY58EqImc2NSY5b%2BiQZnzqOdV%2B79RnbvqlcWg%2Fmk7SaNRyCKP0jJe8zzSHjInCnTZAWxffhZwnfMizKR5eqCI0qziAtmzLocpyXXj3evtVqDebMkPl%2Bwt)\n  - 介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏\n  - 使用TensorFlow 2.0\n## mnist\n- [TensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络)](https:\/\/link.segmentfault.com\/?enc=QsSUwRj8WdLSx60A%2BBIEeg%3D%3D.e%2B19H%2BVWH6S09jrAUwIOOib6GbhQ4zTOnC2FvE2%2BFNILBHtbct7m6wiKWvH15UjlQjqdadX11cwYkOmtV41DCg%3D%3D)\n  - [Github - v4\\_cnn](https:\/\/link.segmentfault.com\/?enc=Q%2Bvm5aAuChDAnr0DsbvBaw%3D%3D.%2F54nd7lyHUxcY%2BLZEdy0bQzgpLnwCz%2FW%2BW6RwtWJjf3SGTV%2FaQXz7Kov18xSf5dUiW9JX2P3A3sAY0PQWIK25vi%2Fs77bIHNWBPH1ZOQjDbGcP6aWXHEJ6pJcc1Z1j%2Bjx)\n  - 介绍了如何搭建CNN网络，准确率达到0.99\n  - 使用TensorFlow 2.0\n- [TensorFlow入门(四) - mnist手写数字识别(制作h5py训练集)](https:\/\/link.segmentfault.com\/?enc=%2F2f4GoYfzapy6HVinn%2BLnA%3D%3D.kj2SJf5ELQxfDEPnajArdbqIG99sWLtliPVqpEYFEmSrOj62Jk2dW3z90RdlkNIPZiS601JItV9GvMmnqbyL8i%2BVE5Atsp4rWpLvizf3mRE%3D)\n  - [Github - make\\_data\\_set](https:\/\/link.segmentfault.com\/?enc=kF5%2BROutx4dB3Juc1o7drg%3D%3D.2tJhiAqOn0FpblPVtCoXKcnIQuKYOGKXh6MviHMpKweNFTum5dnsTxkCkkAPfANeVQXu5cPb990x6DFtNk351hV6VNj8TspepY3BpLoX07xLViO%2FQ1bMIIW8OxqQ%2BTwl)\n  - 介绍了如何使用 numpy 制作 npy 格式的数据集\n  - 介绍了如何使用 h5py 制作 HDF5 格式的数据集\n- [TensorFlow入门(三) - mnist手写数字识别(可视化训练)](https:\/\/link.segmentfault.com\/?enc=jNgHvXdShfzlodCEvYWfoA%3D%3D.IfVnFCTzzbR6I1BAhRr29U8coYBUh4IW52Gbo8HEA7PFpN%2B7ohcoEFt3aDqfr5TdFO7MxyQ72%2FSSevOAXtFtO7EYA1p%2BzmtntDmKKi5U%2FYI%3D)\n  - [Github - mnist\/v3](https:\/\/link.segmentfault.com\/?enc=5J%2BjP3AFH%2FMghDugtXk6vQ%3D%3D.NXQF2uc4vakVn7wqmq4RHc%2FvI9keboVFBsdL%2BVAqRR%2FrZK868zBQs6nhDl5b9%2FKgjs4zAS5c8AqTcKmBL1HCiPEFqhjY1oD6Va1cwgpJxzA%3D)\n  - 介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图\n- [TensorFlow入门(二) - mnist手写数字识别(模型保存加载)](https:\/\/link.segmentfault.com\/?enc=etehP4J%2B2I0Geq%2BSC%2FJETQ%3D%3D.L9%2B8lfB6BaBO3gY2ZWY6P8mfsS5RP7G6wn72NiVafJqOwuW0fhufaQpGj8FfVJqM8JaRIgVjnqwDJdOl4xkNxg%3D%3D)\n  - [Github - mnist\/v2](https:\/\/link.segmentfault.com\/?enc=Q2hmgE3Q31q4mCud4HWuqQ%3D%3D.7vlyZ47n92MO0IxO2rwmHIrpxWK8%2FwHOJ%2FmXU8uwJ3JioYCuuPDAARshJk66ikBsjTkL3KcfEIvqq2HLIyyYDziBdm6qr5XAsmSkjBYHuDs%3D)\n  - 介绍了 TensorFlow 中如何保存训练好的模型\n  - 介绍了如何从某一个模型为起点继续训练\n  - 介绍了模型如何加载使用，传入真实的图片如何识别\n- [TensorFlow入门(一) - mnist手写数字识别(网络搭建)](https:\/\/link.segmentfault.com\/?enc=T%2FE2Ku6OLmGiJAtG10V9xg%3D%3D.7UWG6KIgpijKXi8Qm%2FvgSa1ELT8LiGR8MxbZYAcDlEdeB%2F6iEQlrcyw%2FaLkGs3SVHkS736XXW5pD3tZ%2F%2FfKPDg%3D%3D)\n  - [Github - mnist\/v1](https:\/\/link.segmentfault.com\/?enc=4Sru0MTzWZPQ9gOs2XGNIQ%3D%3D.GB550pI1ykn2hxpOUuuTFfMGTYVDAdaEW8KK77EB6CpHFqgFGKy7GEFebdOzSrFqUQ7nug%2FTIgz6GGR1cG260PAXUi49%2Fv3W8qvrINmp1t0%3D)\n  - 这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。\n  - 包括输入输出、独热编码与损失函数，以及正确率的验证。\n\n[tensorflow](https:\/\/segmentfault.com\/t\/tensorflow)[机器学习](https:\/\/segmentfault.com\/t\/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0)[![](https:\/\/avatar-static.segmentfault.com\/252\/177\/2521771040-54cb53b372821_small)python](https:\/\/segmentfault.com\/t\/python)\n\n赞3\n\n收藏1\n\n分享\n\n阅读 8.9k\n\n[更新于 2020-10-18](https:\/\/segmentfault.com\/a\/1190000019541931\/revision)\n\n举报\n***\n[![头像](https:\/\/avatar-static.segmentfault.com\/417\/773\/417773404-59a19cc559425_huge128)](https:\/\/segmentfault.com\/u\/geektutu)\n\n[极客兔兔](https:\/\/segmentfault.com\/u\/geektutu)\n\n823 声望94 粉丝\n\n专注于新技术的尝试与分享，知识付费。\n\n关注作者\n***\n« 上一篇\n\n下一篇 »\n\n[Pandas 数据分析处理入门实战案例教程](https:\/\/segmentfault.com\/a\/1190000019640941)\n\n▲\n\n3\n\n1\n\n### 引用和评论\n**推荐阅读**\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/417\/773\/417773404-59a19cc559425_big64)Go 语言高性能编程 极客兔兔赞 2阅读 4.3k](https:\/\/segmentfault.com\/a\/1190000038465116?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/328\/674\/3286743184-68aeba39aebb8_big64)实现高选择性底物设计，MIT联手哈佛用生成式AI发现全新蛋白酶切割模式 超神经HyperAI赞 1阅读 3.1k](https:\/\/segmentfault.com\/a\/1190000047537328?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/328\/674\/3286743184-68aeba39aebb8_big64)具身智能资源汇总：机器人学习数据集，在线体验世界建模模型，英伟达\/字节\/小米等最新研究论文 超神经HyperAI阅读 10.2k](https:\/\/segmentfault.com\/a\/1190000047561346?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/228\/006\/2280067195-6978e93971616_big64)基于 YOLOv8 的 100 类中药材智能识别实战 \\[目标检测完整源码\\] 风筝赞 1阅读 1.4k](https:\/\/segmentfault.com\/a\/1190000047535602?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/262\/594\/262594592-61986cbc232bb_big64)微信电脑版4.1.X最新版获取未读消息并在右下角弹出提醒 TANKING阅读 5.8k评论 2](https:\/\/segmentfault.com\/a\/1190000047251654?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/427\/300\/4273004700-60f7c5a323eff_big64)MurmurHash Tips（qbit） qbit阅读 5.4k](https:\/\/segmentfault.com\/a\/1190000020313744?utm_source=sf-similar-article)\n\n[![头像](https:\/\/avatar-static.segmentfault.com\/288\/882\/2888823538-5f681aa01be18_big64)2022KDD论文解读：深度学习订单出餐时间概率预测 喜东东赞 1阅读 5.2k评论 2](https:\/\/segmentfault.com\/a\/1190000042671084?utm_source=sf-similar-article)\n\n**0 条评论**\n\n[得票](https:\/\/segmentfault.com\/a\/1190000019541931?sort=votes)[最新](https:\/\/segmentfault.com\/a\/1190000019541931?sort=newest)\n\n![头像](https:\/\/image-static.segmentfault.com\/317\/931\/3179314346-5f61e47221e07)\n\n提交评论\n\n评论支持部分 Markdown 语法：``**粗体** _斜体_ [链接](http:\/\/example.com) `代码` - 列表 > 引用``。你还可以使用 `@`来通知其他用户。\n\n©2026 极客兔兔\n\n除特别声明外，[作品采用《署名-非商业性使用-禁止演绎 4.0 国际》进行许可](https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/)\n\n[使用 SegmentFault 发布](https:\/\/segmentfault.com\/blogs)\n\n[SegmentFault - 凝聚集体智慧，推动技术进步](https:\/\/segmentfault.com\/)\n\n[服务协议](https:\/\/segmentfault.com\/tos?utm_source=sf-footer)[隐私政策](https:\/\/segmentfault.com\/privacy?utm_source=sf-footer)[浙ICP备15005796号-2](http:\/\/beian.miit.gov.cn\/)[浙公网安备33010602002000号](http:\/\/www.beian.gov.cn\/portal\/registerSystemInfo?recordcode=33010602002000)","attrs_readable_markdown":"> 用最白话的语言，讲解机器学习、神经网络与深度学习  \n> 示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现\n## 相关链接\n- [一篇文章入门 Python](https:\/\/link.segmentfault.com\/?enc=zO40sB6JYWO0CUkCVdslqQ%3D%3D.erm5mTbN70JccY0V1ZfNiOfS%2FvlzHt4vDOs%2Bbij29Ii0GlapYuMxwTgmBLcvO3zw)\n- [机器学习笔试面试题](https:\/\/link.segmentfault.com\/?enc=Rqjl6qSgJgF0v28D20EELA%3D%3D.49JXB6s%2Fcue6%2B1pUi8sf4T1HM3iRSmHarhbYdvrY%2BlpGPUg9gk%2Fl8lEWtKwP48M%2B)，[Github](https:\/\/link.segmentfault.com\/?enc=q91i1qmN60N6Poq3sDGnCg%3D%3D.UCINOkmzdCPCeThxkR%2BPCa5aoSr20ag5TokdDUN%2BVQCU1Bi6atDYbXRwZYYIqXYe)\n- [TensorFlow 2.0 中文文档](https:\/\/link.segmentfault.com\/?enc=NzldmwAnmuqJfKZbg5L0DQ%3D%3D.UPaRQ8aMA1dcYscUyiDu%2BzzluIJHBcXVRPB4FDMsAJCXJfF2T8BWX1LntvX%2FIz3P)，[Github](https:\/\/link.segmentfault.com\/?enc=h2S8L9peI%2FMVYktBI6wysQ%3D%3D.DMC%2BN28Lz7QhbqL80m72eIUw4PfUZJX0csR88olqCcOTOsvnPCV02htIG%2F%2Ft9M0G)\n- [TensorFlow 2.0 图像识别&强化学习实战](https:\/\/link.segmentfault.com\/?enc=ylYxIgxvKgmG1oTC3ynj%2BA%3D%3D.HqX%2FibuVDWOFm0Hr5L2IIt3LEgTYj0oAO2I5T8YtrZG2arQOT3%2F9KTOyDe6RVeqsBIvTM5DwoeiNd15MDFTFcw%3D%3D)，[Github](https:\/\/link.segmentfault.com\/?enc=rdFycR6KsBWOFxj1E6rrFQ%3D%3D.%2F95LbrTMnnaStIhn2JVO%2FyQpDVhQ85E2Hlhfv7mSY2KPP%2FNwLuTpSBYDGsEYh8bPjZk3zcNgBqeWjeDonHCb9g%3D%3D)\n## OpenAI gym\n- [TensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient](https:\/\/link.segmentfault.com\/?enc=1Zl6QZI7bxGrZSY%2BVr0Cvg%3D%3D.qfag0sCR9mlSSlbsuQ44GblRZBmaTTo0lJPQd%2Bctk1tEAzYRo99URx2S3Hol7t5fuhK%2BDjf9wTgR3qsruGoUPg%3D%3D)\n  - [Github - gym\/CartPole-v0-policy-gradient](https:\/\/link.segmentfault.com\/?enc=oB9emap3au%2BbKUUfaT%2F5qw%3D%3D.BWnZrCqluh2cbUZ06RPCbG3kCB5Hv03UMY8r7U5Qp9m1o%2FHas5ouVZMPj3Ygx9S%2B0Yr%2BoaVprIHl1j1YuwJOZOwogmJ8L2wqdheOdoJXFHIBK5rV%2FgI%2B57U1oFPtk28SmFO%2FJZ1VKvJ8amHdnmHMmg%3D%3D)\n  - 介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0\n- [TensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car](https:\/\/link.segmentfault.com\/?enc=63mByzrb9M%2FRkmpAqVHILg%3D%3D.6CXQC2Jmi4wbEu7YT3KK5mNOC2nUoOPe2HpVk5LqFYEYz92YikCZJTGvPCCCyJalkmYTFnJmT7P4Gy4OKp16VA%3D%3D)\n  - [Github - gym\/MountainCar-v0-dqn](https:\/\/link.segmentfault.com\/?enc=WJNlCqmmnEqjUHeQCMVhow%3D%3D.1amA5Uc21lBH754MZTgdmAnwh8qG%2Fux1Ib88U6lZBbpZNqv6Epx2JnQ5F6m7TsoDVgqsjfPmoEoT8FHyELC%2Fw2HHKOGFzE2F3rCwK%2B8RI9LKI5A%2FwD%2B%2FlpDojLJAnODf)\n  - 介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏\n  - Q-Table用神经网络来代替。\n- [TensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym](https:\/\/link.segmentfault.com\/?enc=eODVi5aVheibmzAPf7b5YA%3D%3D.mPK%2FoXCImhFAMGAtYcv3EL8JSsA0fJn%2F1kd39wpVDIreVD9W90u6QeX0a%2BBXvdaq7iK2NQ%2FtExlJHqMKcYiXcQ%3D%3D)\n  - [Github - gym\/MountainCar-v0-q-learning](https:\/\/link.segmentfault.com\/?enc=0UtAsjtjuoULsN3iJZV9qg%3D%3D.ByalkcOoAiq6NRvN%2FXOezAtfxk9Ph17q5C8Ii7Rr6%2B5uV8ojZ4%2BtF2WVFlENLGaGa2J6WH2xs6L75yvzeNQW8hOTmgwp8OhNToPcAK8%2FwVc0G0AjUnS4HuWDpkm56vbe7qhgIYMMVY631dSXt9%2BhdA%3D%3D)\n  - 介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏\n  - 将连续的状态离散化。\n- [TensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game](https:\/\/link.segmentfault.com\/?enc=MuNqKigi1bC3ZZjMqVI%2BwA%3D%3D.aTX6kHUjZXQw%2BrQBBlDblKGljtuowa5aQjCmJeGcy5cpw5JZE%2BNpnt3L7KLq390q4rd%2B9TfsxVPQsdehr3HduQ%3D%3D)\n  - [Github - gym\/CartPole-v0-nn](https:\/\/link.segmentfault.com\/?enc=SUJ7Eh4qv48fOFj5q8OsLw%3D%3D.R%2BNQFxUEMHdCY58EqImc2NSY5b%2BiQZnzqOdV%2B79RnbvqlcWg%2Fmk7SaNRyCKP0jJe8zzSHjInCnTZAWxffhZwnfMizKR5eqCI0qziAtmzLocpyXXj3evtVqDebMkPl%2Bwt)\n  - 介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏\n  - 使用TensorFlow 2.0\n## mnist\n- [TensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络)](https:\/\/link.segmentfault.com\/?enc=QsSUwRj8WdLSx60A%2BBIEeg%3D%3D.e%2B19H%2BVWH6S09jrAUwIOOib6GbhQ4zTOnC2FvE2%2BFNILBHtbct7m6wiKWvH15UjlQjqdadX11cwYkOmtV41DCg%3D%3D)\n  - [Github - v4\\_cnn](https:\/\/link.segmentfault.com\/?enc=Q%2Bvm5aAuChDAnr0DsbvBaw%3D%3D.%2F54nd7lyHUxcY%2BLZEdy0bQzgpLnwCz%2FW%2BW6RwtWJjf3SGTV%2FaQXz7Kov18xSf5dUiW9JX2P3A3sAY0PQWIK25vi%2Fs77bIHNWBPH1ZOQjDbGcP6aWXHEJ6pJcc1Z1j%2Bjx)\n  - 介绍了如何搭建CNN网络，准确率达到0.99\n  - 使用TensorFlow 2.0\n- [TensorFlow入门(四) - mnist手写数字识别(制作h5py训练集)](https:\/\/link.segmentfault.com\/?enc=%2F2f4GoYfzapy6HVinn%2BLnA%3D%3D.kj2SJf5ELQxfDEPnajArdbqIG99sWLtliPVqpEYFEmSrOj62Jk2dW3z90RdlkNIPZiS601JItV9GvMmnqbyL8i%2BVE5Atsp4rWpLvizf3mRE%3D)\n  - [Github - make\\_data\\_set](https:\/\/link.segmentfault.com\/?enc=kF5%2BROutx4dB3Juc1o7drg%3D%3D.2tJhiAqOn0FpblPVtCoXKcnIQuKYOGKXh6MviHMpKweNFTum5dnsTxkCkkAPfANeVQXu5cPb990x6DFtNk351hV6VNj8TspepY3BpLoX07xLViO%2FQ1bMIIW8OxqQ%2BTwl)\n  - 介绍了如何使用 numpy 制作 npy 格式的数据集\n  - 介绍了如何使用 h5py 制作 HDF5 格式的数据集\n- [TensorFlow入门(三) - mnist手写数字识别(可视化训练)](https:\/\/link.segmentfault.com\/?enc=jNgHvXdShfzlodCEvYWfoA%3D%3D.IfVnFCTzzbR6I1BAhRr29U8coYBUh4IW52Gbo8HEA7PFpN%2B7ohcoEFt3aDqfr5TdFO7MxyQ72%2FSSevOAXtFtO7EYA1p%2BzmtntDmKKi5U%2FYI%3D)\n  - [Github - mnist\/v3](https:\/\/link.segmentfault.com\/?enc=5J%2BjP3AFH%2FMghDugtXk6vQ%3D%3D.NXQF2uc4vakVn7wqmq4RHc%2FvI9keboVFBsdL%2BVAqRR%2FrZK868zBQs6nhDl5b9%2FKgjs4zAS5c8AqTcKmBL1HCiPEFqhjY1oD6Va1cwgpJxzA%3D)\n  - 介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图\n- [TensorFlow入门(二) - mnist手写数字识别(模型保存加载)](https:\/\/link.segmentfault.com\/?enc=etehP4J%2B2I0Geq%2BSC%2FJETQ%3D%3D.L9%2B8lfB6BaBO3gY2ZWY6P8mfsS5RP7G6wn72NiVafJqOwuW0fhufaQpGj8FfVJqM8JaRIgVjnqwDJdOl4xkNxg%3D%3D)\n  - [Github - mnist\/v2](https:\/\/link.segmentfault.com\/?enc=Q2hmgE3Q31q4mCud4HWuqQ%3D%3D.7vlyZ47n92MO0IxO2rwmHIrpxWK8%2FwHOJ%2FmXU8uwJ3JioYCuuPDAARshJk66ikBsjTkL3KcfEIvqq2HLIyyYDziBdm6qr5XAsmSkjBYHuDs%3D)\n  - 介绍了 TensorFlow 中如何保存训练好的模型\n  - 介绍了如何从某一个模型为起点继续训练\n  - 介绍了模型如何加载使用，传入真实的图片如何识别\n- [TensorFlow入门(一) - mnist手写数字识别(网络搭建)](https:\/\/link.segmentfault.com\/?enc=T%2FE2Ku6OLmGiJAtG10V9xg%3D%3D.7UWG6KIgpijKXi8Qm%2FvgSa1ELT8LiGR8MxbZYAcDlEdeB%2F6iEQlrcyw%2FaLkGs3SVHkS736XXW5pD3tZ%2F%2FfKPDg%3D%3D)\n  - [Github - mnist\/v1](https:\/\/link.segmentfault.com\/?enc=4Sru0MTzWZPQ9gOs2XGNIQ%3D%3D.GB550pI1ykn2hxpOUuuTFfMGTYVDAdaEW8KK77EB6CpHFqgFGKy7GEFebdOzSrFqUQ7nug%2FTIgz6GGR1cG260PAXUi49%2Fv3W8qvrINmp1t0%3D)\n  - 这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。\n  - 包括输入输出、独热编码与损失函数，以及正确率的验证。","meta_canonical":null}

3. Robots.txt Check

Query:

Response:

4. Spam/Ban Check

Query:

Response:

5. Seen Status Check

ℹ️ Skipped - page is already crawled

📄

INDEXABLE

✅

CRAWLED

5 days ago

🤖

ROBOTS ALLOWED

Page Info Filters

Filter	Status	Condition	Details
HTTP status	PASS	`download_http_code = 200`	HTTP 200
Age cutoff	PASS	`download_stamp > now() - 6 MONTH`	0.2 months ago
History drop	PASS	`isNull(history_drop_reason)`	No drop reason
Spam/ban	PASS	`fh_dont_index != 1 AND ml_spam_score = 0`	ml_spam_score=0
Canonical	PASS	`meta_canonical IS NULL OR = '' OR = src_unparsed`	Not set

Page Details

Property	Value
URL	https://segmentfault.com/a/1190000019541931
Last Crawled	2026-04-11 03:15:04 (5 days ago)
First Indexed	2019-06-21 12:04:06 (6 years ago)
HTTP Status Code	200
Meta Title	python - TensorFlow 2.0 / TF2.0 入门教程实战案例 - 极客兔兔 - SegmentFault 思否
Meta Description	用最白话的语言，讲解机器学习、神经网络与深度学习示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现相关链接一篇文章入门 Python机器学习笔试面试题，Github...
Meta Canonical	null
Boilerpipe Text	用最白话的语言，讲解机器学习、神经网络与深度学习示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现相关链接一篇文章入门 Python 机器学习笔试面试题， Github TensorFlow 2.0 中文文档， Github TensorFlow 2.0 图像识别&强化学习实战， Github OpenAI gym TensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient Github - gym/CartPole-v0-policy-gradient 介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0 TensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car Github - gym/MountainCar-v0-dqn 介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏 Q-Table用神经网络来代替。 TensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym Github - gym/MountainCar-v0-q-learning 介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏将连续的状态离散化。 TensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game Github - gym/CartPole-v0-nn 介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏使用TensorFlow 2.0 mnist TensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络) Github - v4_cnn 介绍了如何搭建CNN网络，准确率达到0.99 使用TensorFlow 2.0 TensorFlow入门(四) - mnist手写数字识别(制作h5py训练集) Github - make_data_set 介绍了如何使用 numpy 制作 npy 格式的数据集介绍了如何使用 h5py 制作 HDF5 格式的数据集 TensorFlow入门(三) - mnist手写数字识别(可视化训练) Github - mnist/v3 介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图 TensorFlow入门(二) - mnist手写数字识别(模型保存加载) Github - mnist/v2 介绍了 TensorFlow 中如何保存训练好的模型介绍了如何从某一个模型为起点继续训练介绍了模型如何加载使用，传入真实的图片如何识别 TensorFlow入门(一) - mnist手写数字识别(网络搭建) Github - mnist/v1 这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。包括输入输出、独热编码与损失函数，以及正确率的验证。
Markdown	[极客兔兔](https://segmentfault.com/blog/geektutu) [极客兔兔](https://segmentfault.com/blog/geektutu) [注册登录](https://segmentfault.com/user/login) [关注博客](https://segmentfault.com/a/1190000019541931)[注册登录](https://segmentfault.com/user/login) [主页](https://segmentfault.com/blog/geektutu) [关于](https://segmentfault.com/blog/geektutu/about) [RSS](https://segmentfault.com/feeds/blog/geektutu) # [TensorFlow 2.0 / TF2.0 入门教程实战案例](https://segmentfault.com/a/1190000019541931) [![头像](https://avatar-static.segmentfault.com/417/773/417773404-59a19cc559425_huge128)极客兔兔](https://segmentfault.com/u/geektutu) [2019-06-21](https://segmentfault.com/a/1190000019541931/revision) 阅读 2 分钟 3 > 用最白话的语言，讲解机器学习、神经网络与深度学习 > 示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现 ## 相关链接 - [一篇文章入门 Python](https://link.segmentfault.com/?enc=zO40sB6JYWO0CUkCVdslqQ%3D%3D.erm5mTbN70JccY0V1ZfNiOfS%2FvlzHt4vDOs%2Bbij29Ii0GlapYuMxwTgmBLcvO3zw) - [机器学习笔试面试题](https://link.segmentfault.com/?enc=Rqjl6qSgJgF0v28D20EELA%3D%3D.49JXB6s%2Fcue6%2B1pUi8sf4T1HM3iRSmHarhbYdvrY%2BlpGPUg9gk%2Fl8lEWtKwP48M%2B)，[Github](https://link.segmentfault.com/?enc=q91i1qmN60N6Poq3sDGnCg%3D%3D.UCINOkmzdCPCeThxkR%2BPCa5aoSr20ag5TokdDUN%2BVQCU1Bi6atDYbXRwZYYIqXYe) - [TensorFlow 2.0 中文文档](https://link.segmentfault.com/?enc=NzldmwAnmuqJfKZbg5L0DQ%3D%3D.UPaRQ8aMA1dcYscUyiDu%2BzzluIJHBcXVRPB4FDMsAJCXJfF2T8BWX1LntvX%2FIz3P)，[Github](https://link.segmentfault.com/?enc=h2S8L9peI%2FMVYktBI6wysQ%3D%3D.DMC%2BN28Lz7QhbqL80m72eIUw4PfUZJX0csR88olqCcOTOsvnPCV02htIG%2F%2Ft9M0G) - [TensorFlow 2.0 图像识别&强化学习实战](https://link.segmentfault.com/?enc=ylYxIgxvKgmG1oTC3ynj%2BA%3D%3D.HqX%2FibuVDWOFm0Hr5L2IIt3LEgTYj0oAO2I5T8YtrZG2arQOT3%2F9KTOyDe6RVeqsBIvTM5DwoeiNd15MDFTFcw%3D%3D)，[Github](https://link.segmentfault.com/?enc=rdFycR6KsBWOFxj1E6rrFQ%3D%3D.%2F95LbrTMnnaStIhn2JVO%2FyQpDVhQ85E2Hlhfv7mSY2KPP%2FNwLuTpSBYDGsEYh8bPjZk3zcNgBqeWjeDonHCb9g%3D%3D) ## OpenAI gym - [TensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient](https://link.segmentfault.com/?enc=1Zl6QZI7bxGrZSY%2BVr0Cvg%3D%3D.qfag0sCR9mlSSlbsuQ44GblRZBmaTTo0lJPQd%2Bctk1tEAzYRo99URx2S3Hol7t5fuhK%2BDjf9wTgR3qsruGoUPg%3D%3D) - [Github - gym/CartPole-v0-policy-gradient](https://link.segmentfault.com/?enc=oB9emap3au%2BbKUUfaT%2F5qw%3D%3D.BWnZrCqluh2cbUZ06RPCbG3kCB5Hv03UMY8r7U5Qp9m1o%2FHas5ouVZMPj3Ygx9S%2B0Yr%2BoaVprIHl1j1YuwJOZOwogmJ8L2wqdheOdoJXFHIBK5rV%2FgI%2B57U1oFPtk28SmFO%2FJZ1VKvJ8amHdnmHMmg%3D%3D) - 介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0 - [TensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car](https://link.segmentfault.com/?enc=63mByzrb9M%2FRkmpAqVHILg%3D%3D.6CXQC2Jmi4wbEu7YT3KK5mNOC2nUoOPe2HpVk5LqFYEYz92YikCZJTGvPCCCyJalkmYTFnJmT7P4Gy4OKp16VA%3D%3D) - [Github - gym/MountainCar-v0-dqn](https://link.segmentfault.com/?enc=WJNlCqmmnEqjUHeQCMVhow%3D%3D.1amA5Uc21lBH754MZTgdmAnwh8qG%2Fux1Ib88U6lZBbpZNqv6Epx2JnQ5F6m7TsoDVgqsjfPmoEoT8FHyELC%2Fw2HHKOGFzE2F3rCwK%2B8RI9LKI5A%2FwD%2B%2FlpDojLJAnODf) - 介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏 - Q-Table用神经网络来代替。 - [TensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym](https://link.segmentfault.com/?enc=eODVi5aVheibmzAPf7b5YA%3D%3D.mPK%2FoXCImhFAMGAtYcv3EL8JSsA0fJn%2F1kd39wpVDIreVD9W90u6QeX0a%2BBXvdaq7iK2NQ%2FtExlJHqMKcYiXcQ%3D%3D) - [Github - gym/MountainCar-v0-q-learning](https://link.segmentfault.com/?enc=0UtAsjtjuoULsN3iJZV9qg%3D%3D.ByalkcOoAiq6NRvN%2FXOezAtfxk9Ph17q5C8Ii7Rr6%2B5uV8ojZ4%2BtF2WVFlENLGaGa2J6WH2xs6L75yvzeNQW8hOTmgwp8OhNToPcAK8%2FwVc0G0AjUnS4HuWDpkm56vbe7qhgIYMMVY631dSXt9%2BhdA%3D%3D) - 介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏 - 将连续的状态离散化。 - [TensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game](https://link.segmentfault.com/?enc=MuNqKigi1bC3ZZjMqVI%2BwA%3D%3D.aTX6kHUjZXQw%2BrQBBlDblKGljtuowa5aQjCmJeGcy5cpw5JZE%2BNpnt3L7KLq390q4rd%2B9TfsxVPQsdehr3HduQ%3D%3D) - [Github - gym/CartPole-v0-nn](https://link.segmentfault.com/?enc=SUJ7Eh4qv48fOFj5q8OsLw%3D%3D.R%2BNQFxUEMHdCY58EqImc2NSY5b%2BiQZnzqOdV%2B79RnbvqlcWg%2Fmk7SaNRyCKP0jJe8zzSHjInCnTZAWxffhZwnfMizKR5eqCI0qziAtmzLocpyXXj3evtVqDebMkPl%2Bwt) - 介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏 - 使用TensorFlow 2.0 ## mnist - [TensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络)](https://link.segmentfault.com/?enc=QsSUwRj8WdLSx60A%2BBIEeg%3D%3D.e%2B19H%2BVWH6S09jrAUwIOOib6GbhQ4zTOnC2FvE2%2BFNILBHtbct7m6wiKWvH15UjlQjqdadX11cwYkOmtV41DCg%3D%3D) - [Github - v4\_cnn](https://link.segmentfault.com/?enc=Q%2Bvm5aAuChDAnr0DsbvBaw%3D%3D.%2F54nd7lyHUxcY%2BLZEdy0bQzgpLnwCz%2FW%2BW6RwtWJjf3SGTV%2FaQXz7Kov18xSf5dUiW9JX2P3A3sAY0PQWIK25vi%2Fs77bIHNWBPH1ZOQjDbGcP6aWXHEJ6pJcc1Z1j%2Bjx) - 介绍了如何搭建CNN网络，准确率达到0.99 - 使用TensorFlow 2.0 - [TensorFlow入门(四) - mnist手写数字识别(制作h5py训练集)](https://link.segmentfault.com/?enc=%2F2f4GoYfzapy6HVinn%2BLnA%3D%3D.kj2SJf5ELQxfDEPnajArdbqIG99sWLtliPVqpEYFEmSrOj62Jk2dW3z90RdlkNIPZiS601JItV9GvMmnqbyL8i%2BVE5Atsp4rWpLvizf3mRE%3D) - [Github - make\_data\_set](https://link.segmentfault.com/?enc=kF5%2BROutx4dB3Juc1o7drg%3D%3D.2tJhiAqOn0FpblPVtCoXKcnIQuKYOGKXh6MviHMpKweNFTum5dnsTxkCkkAPfANeVQXu5cPb990x6DFtNk351hV6VNj8TspepY3BpLoX07xLViO%2FQ1bMIIW8OxqQ%2BTwl) - 介绍了如何使用 numpy 制作 npy 格式的数据集 - 介绍了如何使用 h5py 制作 HDF5 格式的数据集 - [TensorFlow入门(三) - mnist手写数字识别(可视化训练)](https://link.segmentfault.com/?enc=jNgHvXdShfzlodCEvYWfoA%3D%3D.IfVnFCTzzbR6I1BAhRr29U8coYBUh4IW52Gbo8HEA7PFpN%2B7ohcoEFt3aDqfr5TdFO7MxyQ72%2FSSevOAXtFtO7EYA1p%2BzmtntDmKKi5U%2FYI%3D) - [Github - mnist/v3](https://link.segmentfault.com/?enc=5J%2BjP3AFH%2FMghDugtXk6vQ%3D%3D.NXQF2uc4vakVn7wqmq4RHc%2FvI9keboVFBsdL%2BVAqRR%2FrZK868zBQs6nhDl5b9%2FKgjs4zAS5c8AqTcKmBL1HCiPEFqhjY1oD6Va1cwgpJxzA%3D) - 介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图 - [TensorFlow入门(二) - mnist手写数字识别(模型保存加载)](https://link.segmentfault.com/?enc=etehP4J%2B2I0Geq%2BSC%2FJETQ%3D%3D.L9%2B8lfB6BaBO3gY2ZWY6P8mfsS5RP7G6wn72NiVafJqOwuW0fhufaQpGj8FfVJqM8JaRIgVjnqwDJdOl4xkNxg%3D%3D) - [Github - mnist/v2](https://link.segmentfault.com/?enc=Q2hmgE3Q31q4mCud4HWuqQ%3D%3D.7vlyZ47n92MO0IxO2rwmHIrpxWK8%2FwHOJ%2FmXU8uwJ3JioYCuuPDAARshJk66ikBsjTkL3KcfEIvqq2HLIyyYDziBdm6qr5XAsmSkjBYHuDs%3D) - 介绍了 TensorFlow 中如何保存训练好的模型 - 介绍了如何从某一个模型为起点继续训练 - 介绍了模型如何加载使用，传入真实的图片如何识别 - [TensorFlow入门(一) - mnist手写数字识别(网络搭建)](https://link.segmentfault.com/?enc=T%2FE2Ku6OLmGiJAtG10V9xg%3D%3D.7UWG6KIgpijKXi8Qm%2FvgSa1ELT8LiGR8MxbZYAcDlEdeB%2F6iEQlrcyw%2FaLkGs3SVHkS736XXW5pD3tZ%2F%2FfKPDg%3D%3D) - [Github - mnist/v1](https://link.segmentfault.com/?enc=4Sru0MTzWZPQ9gOs2XGNIQ%3D%3D.GB550pI1ykn2hxpOUuuTFfMGTYVDAdaEW8KK77EB6CpHFqgFGKy7GEFebdOzSrFqUQ7nug%2FTIgz6GGR1cG260PAXUi49%2Fv3W8qvrINmp1t0%3D) - 这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。 - 包括输入输出、独热编码与损失函数，以及正确率的验证。 [tensorflow](https://segmentfault.com/t/tensorflow)[机器学习](https://segmentfault.com/t/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0)[![](https://avatar-static.segmentfault.com/252/177/2521771040-54cb53b372821_small)python](https://segmentfault.com/t/python) 赞3 收藏1 分享阅读 8.9k [更新于 2020-10-18](https://segmentfault.com/a/1190000019541931/revision) 举报 * [![头像](https://avatar-static.segmentfault.com/417/773/417773404-59a19cc559425_huge128)](https://segmentfault.com/u/geektutu) [极客兔兔](https://segmentfault.com/u/geektutu) 823 声望94 粉丝专注于新技术的尝试与分享，知识付费。关注作者 * « 上一篇下一篇 » [Pandas 数据分析处理入门实战案例教程](https://segmentfault.com/a/1190000019640941) ▲ 3 1 ### 引用和评论推荐阅读 [![头像](https://avatar-static.segmentfault.com/417/773/417773404-59a19cc559425_big64)Go 语言高性能编程极客兔兔赞 2阅读 4.3k](https://segmentfault.com/a/1190000038465116?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/328/674/3286743184-68aeba39aebb8_big64)实现高选择性底物设计，MIT联手哈佛用生成式AI发现全新蛋白酶切割模式超神经HyperAI赞 1阅读 3.1k](https://segmentfault.com/a/1190000047537328?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/328/674/3286743184-68aeba39aebb8_big64)具身智能资源汇总：机器人学习数据集，在线体验世界建模模型，英伟达/字节/小米等最新研究论文超神经HyperAI阅读 10.2k](https://segmentfault.com/a/1190000047561346?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/228/006/2280067195-6978e93971616_big64)基于 YOLOv8 的 100 类中药材智能识别实战 \[目标检测完整源码\] 风筝赞 1阅读 1.4k](https://segmentfault.com/a/1190000047535602?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/262/594/262594592-61986cbc232bb_big64)微信电脑版4.1.X最新版获取未读消息并在右下角弹出提醒 TANKING阅读 5.8k评论 2](https://segmentfault.com/a/1190000047251654?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/427/300/4273004700-60f7c5a323eff_big64)MurmurHash Tips（qbit） qbit阅读 5.4k](https://segmentfault.com/a/1190000020313744?utm_source=sf-similar-article) [![头像](https://avatar-static.segmentfault.com/288/882/2888823538-5f681aa01be18_big64)2022KDD论文解读：深度学习订单出餐时间概率预测喜东东赞 1阅读 5.2k评论 2](https://segmentfault.com/a/1190000042671084?utm_source=sf-similar-article) 0 条评论 [得票](https://segmentfault.com/a/1190000019541931?sort=votes)[最新](https://segmentfault.com/a/1190000019541931?sort=newest) ![头像](https://image-static.segmentfault.com/317/931/3179314346-5f61e47221e07) 提交评论评论支持部分 Markdown 语法：``粗体 _斜体_ [链接](http://example.com) `代码` - 列表 > 引用``。你还可以使用 `@`来通知其他用户。 ©2026 极客兔兔除特别声明外，[作品采用《署名-非商业性使用-禁止演绎 4.0 国际》进行许可](https://creativecommons.org/licenses/by-nc-nd/4.0/) [使用 SegmentFault 发布](https://segmentfault.com/blogs) [SegmentFault - 凝聚集体智慧，推动技术进步](https://segmentfault.com/) [服务协议](https://segmentfault.com/tos?utm_source=sf-footer)[隐私政策](https://segmentfault.com/privacy?utm_source=sf-footer)[浙ICP备15005796号-2](http://beian.miit.gov.cn/)[浙公网安备33010602002000号](http://www.beian.gov.cn/portal/registerSystemInfo?recordcode=33010602002000)
Readable Markdown	> 用最白话的语言，讲解机器学习、神经网络与深度学习 > 示例基于 TensorFlow 1.4 和 TensorFlow 2.0 实现 ## 相关链接 - [一篇文章入门 Python](https://link.segmentfault.com/?enc=zO40sB6JYWO0CUkCVdslqQ%3D%3D.erm5mTbN70JccY0V1ZfNiOfS%2FvlzHt4vDOs%2Bbij29Ii0GlapYuMxwTgmBLcvO3zw) - [机器学习笔试面试题](https://link.segmentfault.com/?enc=Rqjl6qSgJgF0v28D20EELA%3D%3D.49JXB6s%2Fcue6%2B1pUi8sf4T1HM3iRSmHarhbYdvrY%2BlpGPUg9gk%2Fl8lEWtKwP48M%2B)，[Github](https://link.segmentfault.com/?enc=q91i1qmN60N6Poq3sDGnCg%3D%3D.UCINOkmzdCPCeThxkR%2BPCa5aoSr20ag5TokdDUN%2BVQCU1Bi6atDYbXRwZYYIqXYe) - [TensorFlow 2.0 中文文档](https://link.segmentfault.com/?enc=NzldmwAnmuqJfKZbg5L0DQ%3D%3D.UPaRQ8aMA1dcYscUyiDu%2BzzluIJHBcXVRPB4FDMsAJCXJfF2T8BWX1LntvX%2FIz3P)，[Github](https://link.segmentfault.com/?enc=h2S8L9peI%2FMVYktBI6wysQ%3D%3D.DMC%2BN28Lz7QhbqL80m72eIUw4PfUZJX0csR88olqCcOTOsvnPCV02htIG%2F%2Ft9M0G) - [TensorFlow 2.0 图像识别&强化学习实战](https://link.segmentfault.com/?enc=ylYxIgxvKgmG1oTC3ynj%2BA%3D%3D.HqX%2FibuVDWOFm0Hr5L2IIt3LEgTYj0oAO2I5T8YtrZG2arQOT3%2F9KTOyDe6RVeqsBIvTM5DwoeiNd15MDFTFcw%3D%3D)，[Github](https://link.segmentfault.com/?enc=rdFycR6KsBWOFxj1E6rrFQ%3D%3D.%2F95LbrTMnnaStIhn2JVO%2FyQpDVhQ85E2Hlhfv7mSY2KPP%2FNwLuTpSBYDGsEYh8bPjZk3zcNgBqeWjeDonHCb9g%3D%3D) ## OpenAI gym - [TensorFlow 2.0 (九) - 强化学习70行代码实战 Policy Gradient](https://link.segmentfault.com/?enc=1Zl6QZI7bxGrZSY%2BVr0Cvg%3D%3D.qfag0sCR9mlSSlbsuQ44GblRZBmaTTo0lJPQd%2Bctk1tEAzYRo99URx2S3Hol7t5fuhK%2BDjf9wTgR3qsruGoUPg%3D%3D) - [Github - gym/CartPole-v0-policy-gradient](https://link.segmentfault.com/?enc=oB9emap3au%2BbKUUfaT%2F5qw%3D%3D.BWnZrCqluh2cbUZ06RPCbG3kCB5Hv03UMY8r7U5Qp9m1o%2FHas5ouVZMPj3Ygx9S%2B0Yr%2BoaVprIHl1j1YuwJOZOwogmJ8L2wqdheOdoJXFHIBK5rV%2FgI%2B57U1oFPtk28SmFO%2FJZ1VKvJ8amHdnmHMmg%3D%3D) - 介绍了策略梯度算法(Policy Gradient)来玩 CartPole-v0 - [TensorFlow 2.0 (八) - 强化学习 DQN 玩转 gym Mountain Car](https://link.segmentfault.com/?enc=63mByzrb9M%2FRkmpAqVHILg%3D%3D.6CXQC2Jmi4wbEu7YT3KK5mNOC2nUoOPe2HpVk5LqFYEYz92YikCZJTGvPCCCyJalkmYTFnJmT7P4Gy4OKp16VA%3D%3D) - [Github - gym/MountainCar-v0-dqn](https://link.segmentfault.com/?enc=WJNlCqmmnEqjUHeQCMVhow%3D%3D.1amA5Uc21lBH754MZTgdmAnwh8qG%2Fux1Ib88U6lZBbpZNqv6Epx2JnQ5F6m7TsoDVgqsjfPmoEoT8FHyELC%2Fw2HHKOGFzE2F3rCwK%2B8RI9LKI5A%2FwD%2B%2FlpDojLJAnODf) - 介绍了DQN(Deep Q-Learning)来玩MountainCar-v0游戏 - Q-Table用神经网络来代替。 - [TensorFlow 2.0 (七) - 强化学习 Q-Learning 玩转 OpenAI gym](https://link.segmentfault.com/?enc=eODVi5aVheibmzAPf7b5YA%3D%3D.mPK%2FoXCImhFAMGAtYcv3EL8JSsA0fJn%2F1kd39wpVDIreVD9W90u6QeX0a%2BBXvdaq7iK2NQ%2FtExlJHqMKcYiXcQ%3D%3D) - [Github - gym/MountainCar-v0-q-learning](https://link.segmentfault.com/?enc=0UtAsjtjuoULsN3iJZV9qg%3D%3D.ByalkcOoAiq6NRvN%2FXOezAtfxk9Ph17q5C8Ii7Rr6%2B5uV8ojZ4%2BtF2WVFlENLGaGa2J6WH2xs6L75yvzeNQW8hOTmgwp8OhNToPcAK8%2FwVc0G0AjUnS4HuWDpkm56vbe7qhgIYMMVY631dSXt9%2BhdA%3D%3D) - 介绍了使用Q-Learning(创建Q-Table)来玩MountainCar-v0游戏 - 将连续的状态离散化。 - [TensorFlow 2.0 (六) - 监督学习玩转 OpenAI gym game](https://link.segmentfault.com/?enc=MuNqKigi1bC3ZZjMqVI%2BwA%3D%3D.aTX6kHUjZXQw%2BrQBBlDblKGljtuowa5aQjCmJeGcy5cpw5JZE%2BNpnt3L7KLq390q4rd%2B9TfsxVPQsdehr3HduQ%3D%3D) - [Github - gym/CartPole-v0-nn](https://link.segmentfault.com/?enc=SUJ7Eh4qv48fOFj5q8OsLw%3D%3D.R%2BNQFxUEMHdCY58EqImc2NSY5b%2BiQZnzqOdV%2B79RnbvqlcWg%2Fmk7SaNRyCKP0jJe8zzSHjInCnTZAWxffhZwnfMizKR5eqCI0qziAtmzLocpyXXj3evtVqDebMkPl%2Bwt) - 介绍了使用纯监督学习(神经网络)来玩CartPole-v0游戏 - 使用TensorFlow 2.0 ## mnist - [TensorFlow 2.0 (五) - mnist手写数字识别(CNN卷积神经网络)](https://link.segmentfault.com/?enc=QsSUwRj8WdLSx60A%2BBIEeg%3D%3D.e%2B19H%2BVWH6S09jrAUwIOOib6GbhQ4zTOnC2FvE2%2BFNILBHtbct7m6wiKWvH15UjlQjqdadX11cwYkOmtV41DCg%3D%3D) - [Github - v4\_cnn](https://link.segmentfault.com/?enc=Q%2Bvm5aAuChDAnr0DsbvBaw%3D%3D.%2F54nd7lyHUxcY%2BLZEdy0bQzgpLnwCz%2FW%2BW6RwtWJjf3SGTV%2FaQXz7Kov18xSf5dUiW9JX2P3A3sAY0PQWIK25vi%2Fs77bIHNWBPH1ZOQjDbGcP6aWXHEJ6pJcc1Z1j%2Bjx) - 介绍了如何搭建CNN网络，准确率达到0.99 - 使用TensorFlow 2.0 - [TensorFlow入门(四) - mnist手写数字识别(制作h5py训练集)](https://link.segmentfault.com/?enc=%2F2f4GoYfzapy6HVinn%2BLnA%3D%3D.kj2SJf5ELQxfDEPnajArdbqIG99sWLtliPVqpEYFEmSrOj62Jk2dW3z90RdlkNIPZiS601JItV9GvMmnqbyL8i%2BVE5Atsp4rWpLvizf3mRE%3D) - [Github - make\_data\_set](https://link.segmentfault.com/?enc=kF5%2BROutx4dB3Juc1o7drg%3D%3D.2tJhiAqOn0FpblPVtCoXKcnIQuKYOGKXh6MviHMpKweNFTum5dnsTxkCkkAPfANeVQXu5cPb990x6DFtNk351hV6VNj8TspepY3BpLoX07xLViO%2FQ1bMIIW8OxqQ%2BTwl) - 介绍了如何使用 numpy 制作 npy 格式的数据集 - 介绍了如何使用 h5py 制作 HDF5 格式的数据集 - [TensorFlow入门(三) - mnist手写数字识别(可视化训练)](https://link.segmentfault.com/?enc=jNgHvXdShfzlodCEvYWfoA%3D%3D.IfVnFCTzzbR6I1BAhRr29U8coYBUh4IW52Gbo8HEA7PFpN%2B7ohcoEFt3aDqfr5TdFO7MxyQ72%2FSSevOAXtFtO7EYA1p%2BzmtntDmKKi5U%2FYI%3D) - [Github - mnist/v3](https://link.segmentfault.com/?enc=5J%2BjP3AFH%2FMghDugtXk6vQ%3D%3D.NXQF2uc4vakVn7wqmq4RHc%2FvI9keboVFBsdL%2BVAqRR%2FrZK868zBQs6nhDl5b9%2FKgjs4zAS5c8AqTcKmBL1HCiPEFqhjY1oD6Va1cwgpJxzA%3D) - 介绍了tensorboard的简单用法，包括标量图、直方图以及网络结构图 - [TensorFlow入门(二) - mnist手写数字识别(模型保存加载)](https://link.segmentfault.com/?enc=etehP4J%2B2I0Geq%2BSC%2FJETQ%3D%3D.L9%2B8lfB6BaBO3gY2ZWY6P8mfsS5RP7G6wn72NiVafJqOwuW0fhufaQpGj8FfVJqM8JaRIgVjnqwDJdOl4xkNxg%3D%3D) - [Github - mnist/v2](https://link.segmentfault.com/?enc=Q2hmgE3Q31q4mCud4HWuqQ%3D%3D.7vlyZ47n92MO0IxO2rwmHIrpxWK8%2FwHOJ%2FmXU8uwJ3JioYCuuPDAARshJk66ikBsjTkL3KcfEIvqq2HLIyyYDziBdm6qr5XAsmSkjBYHuDs%3D) - 介绍了 TensorFlow 中如何保存训练好的模型 - 介绍了如何从某一个模型为起点继续训练 - 介绍了模型如何加载使用，传入真实的图片如何识别 - [TensorFlow入门(一) - mnist手写数字识别(网络搭建)](https://link.segmentfault.com/?enc=T%2FE2Ku6OLmGiJAtG10V9xg%3D%3D.7UWG6KIgpijKXi8Qm%2FvgSa1ELT8LiGR8MxbZYAcDlEdeB%2F6iEQlrcyw%2FaLkGs3SVHkS736XXW5pD3tZ%2F%2FfKPDg%3D%3D) - [Github - mnist/v1](https://link.segmentfault.com/?enc=4Sru0MTzWZPQ9gOs2XGNIQ%3D%3D.GB550pI1ykn2hxpOUuuTFfMGTYVDAdaEW8KK77EB6CpHFqgFGKy7GEFebdOzSrFqUQ7nug%2FTIgz6GGR1cG260PAXUi49%2Fv3W8qvrINmp1t0%3D) - 这篇博客介绍了使用 TensorFlow 搭建最简单的神经网络。 - 包括输入输出、独热编码与损失函数，以及正确率的验证。
Shard	66 (laksa)
Root Hash	4631572693029667066
Unparsed URL	com,segmentfault!/a/1190000019541931 s443