Deep Learning Chinese Translation _deep

Deep Learning Chinese Translation

In the help of many netizens and proofreading, the draft slowly became the first draft. Although there are many problems, at least 90% of the content is readable and accurate. As far as possible, we kept the meaning of the original book Deep learning and kept the statement of the original book.

However, our level is limited and we cannot eliminate the variance of many readers. We still need everyone's advice and help to reduce the translation bias.

All you have to do is read, and then sum up your suggestions and issue (preferably not one at a single mention). If you are sure that your proposal does not need to be discussed, you can initiate PR directly.

Corresponding translators: Chapters 1th, 4, 7, 10, 14, 20 and 12.4, 12.5 are the responsibility of the @swordyork 2nd, 5, 8, 11, 15, 18, which is the responsibility of the @liber145 3rd, 6, 9 and the @KevinLee1110 is responsible for 13th, 16, 17, Chapters 19 and 12.1 to 12.3 are for readers who are responsible for the @futianfan

Please download PDF read directly.

This version of the accuracy has been improved, the reader can be based on the Chinese version, the English version as a supplement to reading and learning, but we still suggest that the researchers read the original. Reasons for publishing and open source

This book will be published by the press, but we are not sure of the specific date. So we can first look at the PDF electronic version, after all, technology is changing.

If you think that the Chinese version of the PDF to help you, I hope that in the future when you will be able to support the publication of a genuine paper books. If you think the Chinese version is not good, I hope you can put forward more suggestions. Thank you very much.

The following are the specific reasons for open source: We are not literary workers, not full-time translation. Relying on us alone, can not give today's translation, many netizens have given us valuable suggestions, so open source has helped a lot of busy. Publishers will give us royalties (we do not know how much, maybe 20,000 or so), we are embarrassed to use, after the consultation that the donation is the most appropriate, to all the contributions of the name of the Netizen. PDF electronic version for technical books is very important, at any time need to query, with a paper version of the road is obviously inappropriate. Many foreign technical books have a corresponding electronic version (although not necessarily genuine), while the domestic almost no. Personally think that this is the publishing house or the author that the national quality is not high enough to take the initiative to pay for the knowledge of the realm, so do not want to "leak" electronic version. The times are progressing and we need to change. In particular, the general quality of translation works is not high, to dare to the world first. Deep learning develops too quickly and rapidly, so we hope that we can learn the relevant knowledge earlier. I think the original author open PDF Electronic version also has a similar consideration, that is, read first and then pay. We think the quality of Chinese population is high enough to pay for knowledge. Of course, this is not paid to us, is paid to the publishers, the press to pay the original author. We do not want the Chinese version of the sales due to the existence of PDF electronic version of the decline. Publishers only value back to the copyright in order to introduce more excellent books in the future. Our open source translation precedent will not be a negative case, later there will be more PDF electronic version. Open source is also involved in copyright issues, for copyright reasons, we do not update the first edition of the PDF file, please use the final version of the paper to prevail. Thanks

We have 3 categories of proofreading staff. The person in charge is the corresponding translator. Simple reading, the statement is not fluent or difficult to understand the place to propose changes. In contrast, the Chinese and English corresponding reading, the elimination of the situation of less than the wrong turn.

All proofing suggestions are saved in the Annotations.txt file in each chapter. Chapter leader Simple Reading in English and Chinese the first chapter @swordyork LC, @SiriusXDJ, @corenel, @NeutronT @linzhp @liber145 of the second chapter of linear algebra @SiriusXDJ @badpoem Chapter III @KevinLee1110 @SiriusXDJ @kkpoker of probability and information theory, @Peiyan fourth chapter numerical calculation @swordyork @zhangyafeikimi @hengqujushi Chapter Fifth machine learning Base @liber145 @wheaio, @huangpingchun @fairmiracle, @linzhp the sixth chapter depth Feedforward network @KevinLee1110 david_chow, @linzhp, @sailordiary &nbsp ; In the seventh chapter, the regularization @swordyork     Eighth chapter in the depth model @liber145 @happynoom, @codeVerySlow @huangpingchun chapter Nineth convolution network @KevinLee1 @zhaoyu611, @corenel @zhiding Tenth Chapter sequence modeling: Circular and recursive network @swordyork LC @zhaoyu611, @yinruiqing 11th chapter of Practice Methodology @liber145   &NBSP ; The 12th chapter applies @swordyork, @futianfan   @corenel 13th Chapter Linear factor Model @futianfan @cloudygoose @ZhiweiYang 14th Chapter Self Encoder @swordyork   @Seaball, @huangpingchun the 15th chapter represents learning @liber145 @cnscottzheng   16th chapter the structural probability model in deep learning @futianfan     17th Chapter Monte Carlo method @futianfan   @sailordiary The 18th Chapter @futianfan   @sailordiary to face the function @liber145     19th Chapter, @hengqujushi the 20th chapter depth generation model @swordyork    

We will be in the paper when the official publication of the book, Thank you, formally thanked you to contribute to the students.

Many students have put forward a lot of suggestions, we are listed here.

@tttwwy @tankeco @fairmiracle @GageGao @huangpingchun @MaHongP of @acgtyrant @yanhuibin315 @Buttonwood @ weijy026a @RuiZhang1993 @zymiboxpay @xingkongliang @oisc @tielei @yuduowu @Qingmu -2016 @HC @xiaomingabc @bengordai n @JoyFYan @minoriwww @khty2000 @gump88 @zdx3578 @PassStory @imwebson @wlbksy @roachsinai @Elvinczp Name:yue-daj Iong @9578577 @linzhp @cnscottzheng @germany-zhu @zhangyafeikimi @showgood163 @gump88 the @kangqf @NeutronT to @badpoem @kkpoker @Seaball @wheaio @angrymidiao @ZhiweiYang @corenel @zhaoyu611 the @SiriusXDJ emisxxy flyingfire Vsooda @dfcv24-china

If there are omissions, please be sure to notify us, you can send mail to echo c3dvcmquew9ya0bnbwfpbc5jb20k | Base64-d. This is what we have to be thankful for, so don't be embarrassed. TODO typesetting attention to a variety of issues or suggestions can be issue, recommend the use of Chinese. Due to copyright issues, we can not upload pictures and bib, please forgive me. Due to copyright issues, we are would not upload figures and the bib file. may be used for study purposes and shall not be used for any commercial activity. Thank you. Markdown format

This format is really important, easy to access, but also easy to index. After the initial conversion, the page is generated, see Deeplearningbook-chinese specifically. Note that this conversion does not put the diagram in, nor does it place the graph. The current use of a single script, based on latex file conversions, may change in the future but the principle is not to directly modify the MD file. The required students can modify the script themselves.

Updating .....

From:https://github.com/exacity/deeplearningbook-chinese?f=tt&hmsr=toutiao.io&utm_medium=toutiao.io &utm_source=toutiao.io