Giter Club home page Giter Club logo

pytorch-doc-zh's Introduction

PyTorch 中文文档 & 教程

PyTorch 是一个针对深度学习, 并且使用 GPU 和 CPU 来优化的 tensor library (tensor库)
正在翻译: 2.0 中文版本
最新 英文教程 最新 英文文档 最新 中文翻译

介绍

欢迎任何人参与和完善:一个人可以走的很快,但是一群人却可以走的更远。

组织构建[网站]

第三方站长[网站]

其他补充

PyTorch 官方入口

建议反馈

关于转载

  • 最近有很多人联系我们,关于内容授权问题!
  • 开源是指知识应该重在传播和迭代(而不是禁止别人转载)
  • 不然你TM在GitHub开源,然后又说不让转载,你TM有病吧!
  • 禁止商业化,符合协议规范,备注地址来源,重点: 不需要发邮件给我们申请

赞助我们

微信&支付宝

pytorch-doc-zh's People

Contributors

alexdyysp avatar alexjakin avatar alixmu avatar bellongyan avatar bruce1408 avatar bxuan694 avatar daydaylight avatar dynmi avatar fadegentle avatar fonttian avatar foxerlee avatar gg-yuki avatar heloowird avatar hhxx2015 avatar infdahai avatar jiangzhonglian avatar kevin198528 avatar keyianpai avatar luxinfeng avatar masteryi-0018 avatar pegasus1993 avatar runzhi214 avatar sight007 avatar wizardforcel avatar yachenkang avatar yangkae avatar yearing1017 avatar yportne13 avatar yuange250 avatar zanshuxun avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

pytorch-doc-zh's Issues

torch.optim

torch.optim 有人翻译吗?没人翻译我翻译啦

整体进度 v1.0(校对)

官方奖励(好消息 好消息)

凡事翻译过贡献者,都可以申请得到官方的纪念礼物:
如果有问题的人私聊: QQ 529815144

邮件部分内容:
As a start, let us know where we can ship you some PyTorch swag! 
We have stickers, t-shirts, hoodies, and backpacks – let us know what sizes you need.

官方认领礼物地址如下: #353

认领须知

最新版本 PyTorch v1.2 校验任务认领地址 - #422

提交的时候不要改动文件名称,即使它跟章节标题不一样也不要改,因为文件名和原文的链接是对应的!!!

留言格式:昵称 + QQ + 章节

需要取消认领的也在此留言。

章节 校验者 进度
教程部分 - -
Deep Learning with PyTorch: A 60 Minute Blitz @FontTian 100%
What is PyTorch? @FontTian 100%
Autograd: Automatic Differentiation @FontTian 100%
Neural Networks @FontTian 100%
Training a Classifier @FontTian 100%
Optional: Data Parallelism @FontTian 100%
Data Loading and Processing Tutorial @dyywinner 100%
Learning PyTorch with Examples @wulongjian
Transfer Learning Tutorial @infdahai 100%
Deploying a Seq2Seq Model with the Hybrid Frontend @FontTian 100%
Saving and Loading Models @luxinfeng
What is torch.nn really? @luxinfeng
Finetuning Torchvision Models @luxinfeng
Spatial Transformer Networks Tutorial
Neural Transfer Using PyTorch @AllenZYJ
Adversarial Example Generation
Transfering a Model from PyTorch to Caffe2 and Mobile using ONNX
Chatbot Tutorial @jiangzhonglian 100%
Generating Names with a Character-Level RNN @hijkzzz 100%
Classifying Names with a Character-Level RNN @hijkzzz
Deep Learning for NLP with Pytorch @FontTian 100%
Introduction to PyTorch @FontTian 100%
Deep Learning with PyTorch @FontTian 100%
Word Embeddings: Encoding Lexical Semantics
Sequence Models and Long-Short Term Memory Networks @FontTian 100%
Advanced: Making Dynamic Decisions and the Bi-LSTM CRF @enningxie 100%
Translation with a Sequence to Sequence Network and Attention @FontTian 100%
DCGAN Tutorial @jiangzhonglian 100%
Reinforcement Learning (DQN) Tutorial
Creating Extensions Using numpy and scipy @FontTian 100%
Custom C++ and CUDA Extensions
Extending TorchScript with Custom C++ Operators
Writing Distributed Applications with PyTorch
PyTorch 1.0 Distributed Trainer with Amazon AWS
ONNX Live Tutorial
Loading a PyTorch Model in C++
Using the PyTorch C++ Frontend
文档部分 - -
Autograd mechanics @AlexJakin 100%
Broadcasting semantics @AlexJakin 100%
CUDA semantics @AlexJakin
Extending PyTorch
Frequently Asked Questions
Multiprocessing best practices
Reproducibility @bruce1408
Serialization semantics
Windows FAQ
torch
Tensors @dyywinner
Random sampling
Serialization, Parallelism, Utilities
Pointwise Ops
Reduction Ops
Comparison Ops
Spectral Ops
Other Operations
BLAS and LAPACK Operations
torch.Tensor
Tensor Attributes
Type Info
torch.sparse
torch.cuda
torch.Storage
torch.nn
torch.nn.functional @Originval 100%
torch.nn.init @dyywinner
torch.optim @zonasw
Automatic differentiation package - torch.autograd
Distributed communication package - torch.distributed
Probability distributions - torch.distributions
Torch Script
Multiprocessing package - torch.multiprocessing
torch.utils.bottleneck
torch.utils.checkpoint
torch.utils.cpp_extension
torch.utils.data
torch.utils.dlpack
torch.hub
torch.utils.model_zoo
torch.onnx
Distributed communication package (deprecated) - torch.distributed.deprecated
torchvision Reference
torchvision.datasets
torchvision.models
torchvision.transforms
torchvision.utils

搜索功能没有?

能不能把facebook的文档copy一份,放在你们网页。 没有搜索功能的文档, 基本上是意义不大

Pytorch 1.4 宣传认领

注意:

  1. 部分站点有每天的数量限制,请大家合理安排。
  2. 请全文转载,文章开头附上这个仓库中原文的链接。
站点 贡献者 进度
CSDN @wizardforcel 100%
博客园 @wizardforcel 100%
OSChina
SegmentFault
掘金
简书 @zhangqizky
搜狐号
百家号 @zhangqizky
知乎专栏 @FontTian

留言格式

+   CSDN:@wizardforcel [链接](https://www.baidu.com/s?wd=pytorch%20%E4%B8%AD%E6%96%87%E6%96%87%E6%A1%A3%20site%3Ablog.csdn.net)
+   博客园:@wizardforcel [链接](https://www.baidu.com/s?wd=pytorch%20%E4%B8%AD%E6%96%87%E6%96%87%E6%A1%A3%20site%3Acnblogs.com)

翻译质量

你们好,我是pytorch team的core dev,很激动看到你们在翻译pytorch的doc。我看了一些翻译的文档,对于翻译质量有所担忧。比方 cuda semantics中很多语句并不通顺,有种机翻+小修改的感觉。例如:

释放所有 unused 的缓存内存

但是,分配器管理的未使用内存仍将显示为使用 nvidia-smi

如果你想创建一个与另一个 tensor 相同类型和大小的 tensor,并用一个或零填充它

使用 nn.DataParallel 而不是多处理

所以想请咨询一下,你们有没有计划进行额外的审核和update。

再次感谢你们的付出。

贡献者列表(临时)

标题 翻译 校对
文档部分 - -
介绍 - -
自动求导机制 @那伊抹微笑 @twinkle
广播语义 @kelisiya @twinkle
CUDA 语义 @Chris @twinkle
扩展 PyTorch @那伊抹微笑 @twinkle
多进程的最佳实践 @Feng Fei @twinkle
序列化语义 @Dongyao Hu @twinkle
PACKAGE 参考 - -
torch @那伊抹微笑 @张假飞
torch.Tensor @Sylvester @张假飞
torch.sparse @王帅 @Timor
torch.Storage @fanxing @Timor
torch.nn @VPrincekin @那伊抹微笑 @Yang Shun @Zhu Yansen @woaichipinngguo @buldajs @Norlan @Yunfeng Wang @NIck @Yucong Zhu @Garry Lam @QianFanCe @dabney777 @Alex @SiKai Yao @QiaoXie @laihongchang @噼里啪啦嘣 @yuhai Li @KrokYin @JoinsenQ @clown9804 @悲喜世界
torch.optim @ZengyuanYu @青梅往事
torch.autograd @ZhenLei Xu @青梅往事
torch.distributions @Ye Shuquan @smile_su
torch.multiprocessing @夜神月 @smile_su
torch.distributed @Mu Wu9527 @smile_su
torch.legacy @那伊抹微笑 @smile_su
torch.cuda @zhu1040028623 @smile_su
torch.utils.ffi @zhiming @aleczhang
torch.utils.data @zhiming @aleczhang
torch.utils.model_zoo @zhiming @aleczhang
torch.onnx @Haofan Wang @aleczhang
TORCHVISION 参考 - -
torchvision @那伊抹微笑 @dawenzi123 @LeeGeong @liandongze @咸鱼
教程部分 - -
初学者教程 - -
PyTorch 深度学习: 60 分钟极速入门教程 @VPrincekin @李子文
为以前 Torch 用户提供的 Pytorch 教程 @unknown @片刻
跟着例子学习 PyTorch @yongjay13 @片刻
迁移学习教程 @Sylvester @片刻
数据加载和处理教程 @distant1219 @片刻
Deep Learning for NLP with Pytorch @JingTao
中级教程 - -
Classifying Names with a Character-Level RNN @孙永杰
Generating Names with a Character-Level RNN @孙永杰
Translation with a Sequence to Sequence Network and Attention @EWilsen
Reinforcement Learning (DQN) tutorial @Lisanaaa
Writing Distributed Applications with PyTorch @Sylvester
Spatial Transformer Networks Tutorial
高级教程 - -
Neural Transfer with PyTorch
Creating extensions using numpy and scipy
Transfering a model from PyTorch to Caffe2 and Mobile using ONNX
Custom C extensions for pytorch

注:斜体代表未完成

Pytorch 官方翻译奖励-认领中

凡事翻译过贡献者,都可以申请得到官方的纪念礼物:
如果有问题的人私聊: QQ 529815144

邮件部分内容:
As a start, let us know where we can ship you some PyTorch swag! 
We have stickers, t-shirts, hoodies, and backpacks – let us know what sizes you need.

image

留下联系方式:
QQ: 529815144 (记得先加我QQ)
收货人: 片刻
手机号码: 131xxxx0490
所在地区: 北京市xxx区xxx镇
详细地址: xxx小区xxxx

重点注意: 选择你喜欢礼物和型号

由于是贡献者,请留下你参与贡献的记录,例如:
(如果没找到,也可以留下 pull resquests 就行 )
image

关于出现冲突

如果你的版本不是仓库最新版本的后继,就会出现这种情况。

所以每次改动文件之前,或者 commit 之后,需要 git pull 来拉取仓库最新的版本。

如果出现冲突,需要自行解决。搜索 <<< 符号,两个版本选一个即可。

详细操作请见相关文档。

blitz_cifar10_tutorial.md精度问题

你好,感谢辛苦付出。
想请问下,训练结束后是不是需要保存模型,再加载它用来训练呢?
我按照教程的话,直接训练,训完测试,精度就一直是30%左右,也就是“盲猜“。
我改了一下,还是不行哎。我这样改的:

    # 读取之前保存的网络模型参数
    checkpoint = torch.load("./outputs/weights/3.pth")
    net.load_state_dict(checkpoint)
    # 使用测试数据测试网络  检查网络是否学到了一些东西。
    test(net)

希望解惑

网站的翻译不准确

http://pytorch.apachecn.org 上的

PyTorch 是第一个进入 Python 版本的深度学习框架.
我们正在进行早期版本的测试. 期待一些商业投资.

我觉得

Expect some adventures.

意思是PyTorch还有些Bug,使用PyTorch就像冒险一样会遇到一些挑战。不是 期待一些商业投资. 😂

【已完成】整体进度 v1.0(翻译)

章节 译者 进度
教程部分 - -
Deep Learning with PyTorch: A 60 Minute Blitz @bat67 100%
What is PyTorch? @bat67 100%
Autograd: Automatic Differentiation @bat67 100%
Neural Networks @bat67 100%
Training a Classifier @bat67 100%
Optional: Data Parallelism @bat67 100%
Data Loading and Processing Tutorial @yportne13 100%
Learning PyTorch with Examples @bat67 100%
Transfer Learning Tutorial @jiangzhonglian 100%
Deploying a Seq2Seq Model with the Hybrid Frontend @cangyunye 100%
Saving and Loading Models @bruce1408 100%
What is torch.nn really? @lhc741 100%
Finetuning Torchvision Models @ZHHAYO 100%
Spatial Transformer Networks Tutorial @PEGASUS1993 100%
Neural Transfer Using PyTorch @bdqfork 100%
Adversarial Example Generation @cangyunye 100%
Transfering a Model from PyTorch to Caffe2 and Mobile using ONNX @PEGASUS1993 100%
Chatbot Tutorial @a625687551 100%
Generating Names with a Character-Level RNN @hhxx2015 100%
Classifying Names with a Character-Level RNN @hhxx2015 100%
Deep Learning for NLP with Pytorch @bruce1408 100%
Introduction to PyTorch @guobaoyo 100%
Deep Learning with PyTorch @bdqfork 100%
Word Embeddings: Encoding Lexical Semantics @sight007 100%
Sequence Models and Long-Short Term Memory Networks @ETCartman 100%
Advanced: Making Dynamic Decisions and the Bi-LSTM CRF @apachecn 100%
Translation with a Sequence to Sequence Network and Attention @mengfu188 100%
DCGAN Tutorial @wangshuai9517 100%
Reinforcement Learning (DQN) Tutorial @friedhelm739 100%
Creating Extensions Using numpy and scipy @cangyunye 100%
Custom C++ and CUDA Extensions @P3n9W31 100%
Extending TorchScript with Custom C++ Operators @apachecn 100%
Writing Distributed Applications with PyTorch @firdameng 100%
PyTorch 1.0 Distributed Trainer with Amazon AWS @yportne13 100%
ONNX Live Tutorial @PEGASUS1993 100%
Loading a PyTorch Model in C++ @talengu 100%
Using the PyTorch C++ Frontend @solerji 100%
文档部分 - -
Autograd mechanics @PEGASUS1993 100%
Broadcasting semantics @PEGASUS1993 100%
CUDA semantics @jiangzhonglian 100%
Extending PyTorch @PEGASUS1993 100%
Frequently Asked Questions @PEGASUS1993 100%
Multiprocessing best practices @cvley 100%
Reproducibility @apachecn 100%
Serialization semantics @yuange250 100%
Windows FAQ @PEGASUS1993 100%
torch @infdahai 100%
Tensors @infdahai
Random sampling @apachecn 100%
Serialization, Parallelism, Utilities @apachecn 100%
Pointwise Ops @apachecn 100%
Reduction Ops @apachecn 100%
Comparison Ops @apachecn 100%
Spectral Ops @apachecn 100%
Other Operations @apachecn 100%
BLAS and LAPACK Operations @apachecn 100%
torch.Tensor @hijkzzz 100%
Tensor Attributes @yuange250 100%
Type Info @PEGASUS1993 100%
torch.sparse @hijkzzz 100%
torch.cuda @bdqfork 100%
torch.Storage @yuange250 100%
torch.nn @gongel 100%
torch.nn.functional @hijkzzz 100%
torch.nn.init @GeneZC 100%
torch.optim @apachecn 100%
Automatic differentiation package - torch.autograd @gfjiangly 100%
Distributed communication package - torch.distributed @univeryinli 100%
Probability distributions - torch.distributions @hijkzzz 100%
Torch Script @keyianpai 100%
Multiprocessing package - torch.multiprocessing @hijkzzz 100%
torch.utils.bottleneck @belonHan 100%
torch.utils.checkpoint @belonHan 100%
torch.utils.cpp_extension @belonHan 100%
torch.utils.data @BXuan694 100%
torch.utils.dlpack @kunwuz 100%
torch.hub @kunwuz 100%
torch.utils.model_zoo @BXuan694 100%
torch.onnx @guobaoyo 100%
Distributed communication package (deprecated) - torch.distributed.deprecated @luxinfeng 100%
torchvision Reference @BXuan694 100%
torchvision.datasets @BXuan694 100%
torchvision.models @BXuan694 100%
torchvision.transforms @BXuan694 100%
torchvision.utils @BXuan694 100%

blitz/neural_networks_tutorial 例子有问题

class Net(nn.Module):

def __init__(self):
    super(Net, self).__init__()
    # 卷积层 '1'表示输入图片为单通道, '6'表示输出通道数, '5'表示卷积核为5*5
    # 核心
    self.conv1 = nn.Conv2d(1, 6, 5)
    self.conv2 = nn.Conv2d(6, 16, 5)  **#这里 如果是按照输入数据的话,应该是 1,16,5**
    # 仿射层/全连接层: y = Wx + b
    self.fc1 = nn.Linear(16 * 5 * 5, 120)
    self.fc2 = nn.Linear(120, 84)
    self.fc3 = nn.Linear(84, 10)

target = Variable(torch.arange(1, 11)) **# 这里应该是torch.randn(1,10)吧。 **

不更改这两处,例子会报错。译文没错,官网文档上也是这样的代码。
可以验证下。

公式显示出错,如何修正?

目前该版本支持 公式

  • 一句话中间的公式写法: $\delta$
  • 独立换行的公式写法:$$L_{out}=floor((L_{in}+2*padding-dilation*(kernerl\_size-1)-1)/stride+1)$$

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.