singa-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From wang...@apache.org
Subject [1/2] incubator-singa git commit: SINGA-352 Synchronize Chinese docs with English ones
Date Sat, 30 Jun 2018 13:39:57 GMT
Repository: incubator-singa
Updated Branches:
  refs/heads/master c8df172c7 -> e248e447b


SINGA-352 Synchronize Chinese docs with English ones


Project: http://git-wip-us.apache.org/repos/asf/incubator-singa/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-singa/commit/f7ff0be7
Tree: http://git-wip-us.apache.org/repos/asf/incubator-singa/tree/f7ff0be7
Diff: http://git-wip-us.apache.org/repos/asf/incubator-singa/diff/f7ff0be7

Branch: refs/heads/master
Commit: f7ff0be7d920e9fdecdb03e59f4008ea0ca4f130
Parents: 9b06c29
Author: Wentong-DST <liwentongnus@gmail.com>
Authored: Fri Jun 29 14:05:24 2018 +0800
Committer: Wentong-DST <liwentongnus@gmail.com>
Committed: Fri Jun 29 14:05:24 2018 +0800

----------------------------------------------------------------------
 doc/zh/develop/schedule.rst | 53 ++++++++++++++++++++--------------------
 doc/zh/downloads.md         | 36 ++++++++++++++++++++++++---
 doc/zh/index.rst            | 29 ++++++----------------
 3 files changed, 67 insertions(+), 51 deletions(-)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/f7ff0be7/doc/zh/develop/schedule.rst
----------------------------------------------------------------------
diff --git a/doc/zh/develop/schedule.rst b/doc/zh/develop/schedule.rst
index bdd821f..51e96e1 100644
--- a/doc/zh/develop/schedule.rst
+++ b/doc/zh/develop/schedule.rst
@@ -22,45 +22,44 @@
 .. csv-table::
   :header: "版本","模块","特性"
 
-  "v0.1 2015 九月     ","神经网络               ","前向传播神经网络, 包括
CNN, MLP                                                                     "
-  "                   ","                             ","类 RBM 模型, 包括 RBM     
                                                                                 "
-  "                   ","                             ","循环神经网络, 包括标准
RNN                                                                    "
-  "                   ","架构                 ","在单节点运行一个工作组  (包括划分)
                                                              "
+  "v0.1 2015 九月     ","神经网络                     ","前向传播神经网络,
包括 CNN, MLP"
+  "                   ","                             ","类 RBM 模型, 包括 RBM"
+  "                   ","                             ","循环神经网络, 包括标准
RNN"
+  "                   ","架构                         ","在单节点运行一个工作组
(包括划分)"
   "                   ","                             ","在单节点运行多个工作组,
用 `Hogwild <http://www.eecs.berkeley.edu/~brecht/papers/hogwildTR.pdf>`_     "
   "                   ","                             ","分布式 Hogwild"
   "                   ","                             ","跨多节点运行多个工作组
, 如 `Downpour <http://papers.nips.cc/paper/4687-large-scale-distritbuted-deep-networks>`_"
   "                   ","                             ","All-Reduce 训练架构如 `DeepImage
<http://arxiv.org/abs/1501.02876>`_                                "
-  "                   ","                             ","服务器间负载均衡       
                                                                                   "
-  "                   ","失败恢复             ","检查点和恢复                 
                                                                            "
-  "                   ","工具                        ","用 GNU 自动工具安装    
                                                                               "
-  "v0.2 2016 一月      ","神经网络               ","前向传播神经网络, 包括
AlexNet, cuDNN 层, 工具                                                  "
-  "                   ","                             ","循环神经网络, 包括 GRU 层和
BPTT                                                               "
-  "                   ","                             ","模型划分和混合划分    
                                                                           "
-  "                   ","工具                        ","融合 Mesos 资源管理     
                                                                "
+  "                   ","                             ","服务器间负载均衡"
+  "                   ","失败恢复                     ","检查点和恢复"
+  "                   ","工具                         ","用 GNU 自动工具安装"
+  "v0.2 2016 一月     ","神经网络                     ","前向传播神经网络,
包括 AlexNet, cuDNN 层, 工具"
+  "                   ","                             ","循环神经网络, 包括 GRU 层和
BPTT"
+  "                   ","                             ","模型划分和混合划分"
+  "                   ","工具                         ","融合 Mesos 资源管理"
   "                   ","                             ","准备部署 Docker images"
   "                   ","                             ","可视化神经网络和调试信息"
-  "                   ","绑定                      ","主要组件 Python 绑定"
+  "                   ","绑定                         ","主要组件 Python 绑定"
   "                   ","GPU                          ","单节点多个 GPU "
-  "v0.3 2016 四月    ","GPU                          ","多个节点, 每个包含多个
GPU"
+  "v0.3 2016 四月     ","GPU                          ","多个节点, 每个包含多个
GPU"
   "                   ","                             ","用 GPU 和 CPU 混合训练  `CcT
<http://arxiv.org/abs/1504.04343>`_"
   "                   ","                             ","支持 cuDNN v4 "
-  "                   ","安装                 ","删除 ZeroMQ, CZMQ 依赖, 单节点训练
zookeeper"
-  "                   ","优化器                      ","添加新的 SGD 优化器,包括
Adam, AdamMax 和 AdaDelta"
-  "                   ","绑定                      ","增强 Python 绑定训练"
-  "v1.0 2016 九月      ","模型抽象      ","Tensor 基于线性代数, 神经网络和随机运算"
+  "                   ","安装                         ","删除 ZeroMQ, CZMQ 依赖, 单节点训练
zookeeper"
+  "                   ","优化器                       ","添加新的 SGD 优化器,包括
Adam, AdamMax 和 AdaDelta"
+  "                   ","绑定                         ","增强 Python 绑定训练"
+  "v1.0 2016 九月     ","模型抽象                     ","Tensor 基于线性代数,
神经网络和随机运算"
   "                   ","                             ","分布式参数更新优化器"
-  "                   ","硬件                     ","使用 Cuda 和 Cudnn for Nvidia GPU"
+  "                   ","硬件                         ","使用 Cuda 和 Cudnn for Nvidia
GPU"
   "                   ","                             ","使用 OpenCL for AMD GPU 及其他设备"
-  "                   ","跨平台               ","从 Linux 扩展到 MacOS"
+  "                   ","跨平台                       ","从 Linux 扩展到 MacOS"
   "                   ","                             ","大型图像模型, 例如, `VGG
<https://arxiv.org/pdf/1409.1556.pdf>`_ 和 `Residual Net <http://arxiv.org/abs/1512.03385>`_"
-  "v1.1 2017 一月      ","模型库                    ","GoogleNet; 医疗健康模型"
-  "                   ","Caffe 转换器              ","使用 SINGA 训练模型, 从 caffe
proto 文件配置"
-  "                   ","模型组件             ","添加 concat 和 slice 层; 接受多个输入网络"
-  "                   ","编译和安装 ","Windows 支持"
+  "v1.1 2017 一月     ","模型库                       ","GoogleNet; 医疗健康模型"
+  "                   ","Caffe 转换器                 ","使用 SINGA 训练模型, 从
caffe proto 文件配置"
+  "                   ","模型组件                     ","添加 concat 和 slice 层;
接受多个输入网络"
+  "                   ","编译和安装                   ","Windows 支持"
   "                   ","                             ","通过与 protobuf 和 openblas
一起编译 SINGA 简化安装"
   "                   ","                             ","用 Jenkins 自动生成 python
wheel"
   "                   ","                             ","从 Debian packages 安装 SINGA"
-  "v1.2 2017 十月      ","Numpy API                    ","依照 numpy API 实现 PySINGA
的 tensor 模块的函数"
-  "                   ","分布式训练         ","从 V0.3 移植分布式训练框架"
-  "                   ","内存优化          ","用新内存池替代 CNMEM 以减少内存消耗"
-  "                   ","执行优化       ","执行调度及运行时优化"
+  "v1.2 2018 六月     ","AutoGrad                     ","后向传播 AutoGrad"
+  "                   ","Python 3                     ","PySinga 支持 Python 3"
+  "                   ","模型                         ","添加流行模型, 包括 VGG,
ResNet, DenseNet, InceptionNet"

http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/f7ff0be7/doc/zh/downloads.md
----------------------------------------------------------------------
diff --git a/doc/zh/downloads.md b/doc/zh/downloads.md
index 8e174f6..819fd34 100644
--- a/doc/zh/downloads.md
+++ b/doc/zh/downloads.md
@@ -1,11 +1,37 @@
 ## 下载 SINGA
 
-* 最新的代码: 请从 [Github](https://github.com/apache/incubator-singa) 的 dev branch
进行 clone
+* 要验证下载的tar.gz文件,请下载KEY和ASC文件,然后执行以下命令
+ 
+        % gpg --import KEYS
+        % gpg --verify downloaded_file.asc downloaded_file
+
+  您还可以检查SHA512或MD5值以查看下载是否已完成
+
+* v1.2.0 (2018年6月6日):
+    * [Apache SINGA 1.2.0](http://www.apache.org/dyn/closer.cgi/incubator/singa/1.2.0/apache-singa-incubating-1.2.0.tar.gz)
+      [\[SHA512\]](https://www.apache.org/dist/incubator/singa/1.2.0/apache-singa-incubating-1.2.0.tar.gz.sha512)
+      [\[KEYS\]](https://www.apache.org/dist/incubator/singa/1.2.0/KEYS)
+      [\[ASC\]](https://www.apache.org/dist/incubator/singa/1.2.0/apache-singa-incubating-1.2.0.tar.gz.asc)
+    * [发行说明 1.2.0](releases/RELEASE_NOTES_1.2.0.html)
+    * 新功能和主要更新,
+        * 实现 autograd (目前支持MLP模型)
+        * 升级 PySinga 以支持 Python 3
+        * 为 Tensor 类添加 stride field
+        * 讲 cuDNN 从 V5 升级到 V7
+        * 为 ImageNet 分类任务添加 VGG, Inception V4, ResNet, 及 DenseNet 模型
+        * 为 conda 包创建别名
+        * 完善中文文档
+        * 添加在 Windows 上运行 Singa 的说明
+        * 更新编译, CI
+        * 修复一些错误
+
+
 
 * v1.1.0 (2017年2月12日):
      * [Apache SINGA 1.1.0](http://www.apache.org/dyn/closer.cgi/incubator/singa/1.1.0/apache-singa-incubating-1.1.0.tar.gz)
-      [\[MD5\]](https://dist.apache.org/repos/dist/release/incubator/singa/1.1.0/apache-singa-incubating-1.1.0.tar.gz.md5)
-      [\[KEYS\]](https://dist.apache.org/repos/dist/release/incubator/singa/1.1.0/KEYS)
+      [\[MD5\]](https://archive.apache.org/dist/incubator/singa/1.1.0/apache-singa-incubating-1.1.0.tar.gz.md5)
+      [\[KEYS\]](https://archive.apache.org/dist/incubator/singa/1.1.0/KEYS)
+      [\[ASC\]](https://archive.apache.org/dist/incubator/singa/1.1.0/apache-singa-incubating-1.1.0.tar.gz.asc)
     * [发行说明 1.1.0](releases/RELEASE_NOTES_1.1.0.html)
     * 新功能和主要更新,
         * 创建 Docker 镜像(CPU和GPU版本)
@@ -29,6 +55,7 @@
     * [Apache SINGA 1.0.0](https://archive.apache.org/dist/incubator/singa/1.0.0/apache-singa-incubating-1.0.0.tar.gz)
       [\[MD5\]](https://archive.apache.org/dist/incubator/singa/1.0.0/apache-singa-incubating-1.0.0.tar.gz.md5)
       [\[KEYS\]](https://archive.apache.org/dist/incubator/singa//1.0.0/KEYS)
+      [\[ASC\]](https://archive.apache.org/dist/incubator/singa/1.0.0/apache-singa-incubating-1.0.0.tar.gz.asc)
     * [发行说明 1.0.0](releases/RELEASE_NOTES_1.0.0.html)
     * 新功能和主要更新,
         * 用于支持更多机器学习模型的张量抽象.
@@ -48,6 +75,7 @@
     * [Apache SINGA 0.3.0](https://archive.apache.org/dist/incubator/singa/0.3.0/apache-singa-incubating-0.3.0.tar.gz)
       [\[MD5\]](https://archive.apache.org/dist/incubator/singa/0.3.0/apache-singa-incubating-0.3.0.tar.gz.md5)
       [\[KEYS\]](https://archive.apache.org/dist/incubator/singa/0.3.0/KEYS)
+      [\[ASC\]](https://archive.apache.org/dist/incubator/singa/0.3.0/apache-singa-incubating-0.3.0.tar.gz.asc)
     * [发行说明 0.3.0](releases/RELEASE_NOTES_0.3.0.html)
     * 新功能和主要更新,
         * [在 GPU 集群上](v0.3.0/gpu.html) 可以在 GPU 集群上训练深度学习模型.
@@ -65,6 +93,7 @@
     * [Apache SINGA 0.2.0](https://archive.apache.org/dist/incubator/singa/0.2.0/apache-singa-incubating-0.2.0.tar.gz)
       [\[MD5\]](https://archive.apache.org/dist/incubator/singa/0.2.0/apache-singa-incubating-0.2.0.tar.gz.md5)
       [\[KEYS\]](https://archive.apache.org/dist/incubator/singa/0.2.0/KEYS)
+      [\[ASC\]](https://archive.apache.org/dist/incubator/singa/0.2.0/apache-singa-incubating-0.2.0.tar.gz.asc)
     * [发行说明 0.2.0](releases/RELEASE_NOTES_0.2.0.html)
     * 新功能和主要更新,
         * [在 GPU 上进行训练](v0.2.0/gpu.html) 可以在具有多个 GPU 卡的单个节点上对复杂模型进行训练.
@@ -84,6 +113,7 @@
     * [Apache SINGA 0.1.0](https://archive.apache.org/dist/incubator/singa/apache-singa-incubating-0.1.0.tar.gz)
       [\[MD5\]](https://archive.apache.org/dist/incubator/singa/apache-singa-incubating-0.1.0.tar.gz.md5)
       [\[KEYS\]](https://archive.apache.org/dist/incubator/singa/KEYS)
+      [\[ASC\]](https://archive.apache.org/dist/incubator/singa/apache-singa-incubating-0.1.0.tar.gz.asc)
     * [Amazon EC2 image](https://console.aws.amazon.com/ec2/v2/home?region=ap-southeast-1#LaunchInstanceWizard:ami=ami-b41001e6)
     * [发行说明 0.1.0](releases/RELEASE_NOTES_0.1.0.html)
     * 主要功能包括,

http://git-wip-us.apache.org/repos/asf/incubator-singa/blob/f7ff0be7/doc/zh/index.rst
----------------------------------------------------------------------
diff --git a/doc/zh/index.rst b/doc/zh/index.rst
index 48574ee..ec608cd 100644
--- a/doc/zh/index.rst
+++ b/doc/zh/index.rst
@@ -19,7 +19,6 @@
 SINGA 中文文档
 ==============
 
-
 最新动态
 -----------
 
@@ -57,32 +56,20 @@ SINGA 中文文档
 
 SINGA 入门
 ---------------
-* 无需安装,直接使用 `AWS <https://aws.amazon.com/marketplace/pp/B01NAUAWZW>`_
or via `Docker <https://hub.docker.com/r/nusdbsystem/singa/>`_.
+* 安装SINGA,通过 `conda <../en/docs/installation.html>`_, apt-get, 或者源码.
 
-* 安装SINGA,通过 `conda <../en/docs/installation.html#from-conda>`_, `Debian
packages <../en/docs/installation.html#from-debian-package>`_  或者 `源码 <../en/docs/installation.html#from-source>`_.
+* 在 `AWS <https://aws.amazon.com/marketplace/pp/B01NAUAWZW>`_ 上, 或者通过 `Docker
<https://hub.docker.com/r/nusdbsystem/singa/>`_ 试用 SINGA.
 
-* 更多操作和实例,参考 `Jupyter notebooks <http://nbviewer.jupyter.org/github/apache/incubator-singa/blob/master/doc/en/docs/notebook/index.ipynb>`_

+* 更多操作和实例,参考 `Jupyter notebooks <http://nbviewer.jupyter.org/github/apache/incubator-singa/blob/master/doc/en/docs/notebook/index.ipynb>`_
以及 `模型库页面  <./docs/model_zoo/index.html>`_ 了解更多例子.
 
-.. |logo1| image:: _static/github.png
-   :scale: 100%
-   :align: middle
-   :target: https://github.com/apache/incubator-singa
-.. |logo2| image:: _static/awsmp.gif 
-   :scale: 100%
-   :align: middle
-   :target: https://aws.amazon.com/marketplace/seller-profile?id=5bcac385-12c4-4802-aec7-351e09b77b4c
-.. |logo3| image:: _static/docker.png
-   :scale: 50%
-   :align: middle
-   :target: https://hub.docker.com/r/nusdbsystem/singa/
-.. |logo4| image:: _static/jupyter.png
+.. |logo| image:: _static/jupyter.png
    :scale: 25%
    :align: middle
    :target: http://nbviewer.jupyter.org/github/apache/incubator-singa/blob/master/doc/en/docs/notebook/index.ipynb
 
-+---------+---------+---------+---------+
-| |logo1| | |logo2| | |logo3| | |logo4| |
-+---------+---------+---------+---------+
++---------+
+| |logo|  |
++---------+
 
 文档
 -------------
@@ -117,8 +104,8 @@ SINGA 入门
 .. toctree::
    :hidden:
 
-   downloads
    docs/index
+   downloads
 
 .. toctree::
    :hidden:


Mime
View raw message