README.md 5.1 KB
Newer Older
Z
zhangjinchao01 已提交
1 2
# PaddlePaddle

C
Cheerego 已提交
3
English | [简体中文](./README_cn.md)
G
gangliao 已提交
4

G
gangliao 已提交
5
[![Build Status](https://travis-ci.org/PaddlePaddle/Paddle.svg?branch=develop)](https://travis-ci.org/PaddlePaddle/Paddle)
S
shanyi15 已提交
6 7
[![Documentation Status](https://img.shields.io/badge/docs-latest-brightgreen.svg?style=flat)](http://www.paddlepaddle.org/documentation/docs/en/1.4/beginners_guide/index_en.html)
[![Documentation Status](https://img.shields.io/badge/中文文档-最新-brightgreen.svg)](http://www.paddlepaddle.org/documentation/docs/zh/1.4/beginners_guide/index_cn.html)
G
gangliao 已提交
8
[![Release](https://img.shields.io/github/release/PaddlePaddle/Paddle.svg)](https://github.com/PaddlePaddle/Paddle/releases)
L
liaogang 已提交
9 10
[![License](https://img.shields.io/badge/license-Apache%202-blue.svg)](LICENSE)

G
gangliao 已提交
11
Welcome to the PaddlePaddle GitHub.
E
emailweixu 已提交
12

Z
zhangjinchao01 已提交
13 14 15 16 17
PaddlePaddle (PArallel Distributed Deep LEarning) is an easy-to-use,
efficient, flexible and scalable deep learning platform, which is originally
developed by Baidu scientists and engineers for the purpose of applying deep
learning to many products at Baidu.

G
gangliao 已提交
18
Our vision is to enable deep learning for everyone via PaddlePaddle.
G
gangliao 已提交
19
Please refer to our [release announcement](https://github.com/PaddlePaddle/Paddle/releases) to track the latest feature of PaddlePaddle.
G
gangliao 已提交
20

21
### Latest PaddlePaddle Release: [Fluid 1.5.0](https://github.com/PaddlePaddle/Paddle/tree/release/1.5)
X
Xin Pan 已提交
22 23 24 25 26 27
### Install Latest Stable Release:
```
# Linux CPU
pip install paddlepaddle
# Linux GPU cuda9cudnn7
pip install paddlepaddle-gpu
28 29
# Linux GPU cuda10cudnn7
pip install paddlepaddle-gpu==1.5.0.post107
X
Xin Pan 已提交
30
# Linux GPU cuda8cudnn7
31
pip install paddlepaddle-gpu==1.5.0.post87
X
Xin Pan 已提交
32 33 34

# For installation on other platform, refer to http://paddlepaddle.org/
```
A
adaxi123 已提交
35
Now our developers could acquire Tesla V100 online computing resources for free. If you create a program by AI Studio, you would obtain 12 hours to train models online per day. If you could insist on that for five consecutive days, then you would own extra 48 hours. [Click here to start](http://ai.baidu.com/support/news?action=detail&id=981).
X
Xin Pan 已提交
36

Z
zhangjinchao01 已提交
37 38 39 40
## Features

- **Flexibility**

G
gangliao 已提交
41 42 43 44
    PaddlePaddle supports a wide range of neural network architectures and
    optimization algorithms. It is easy to configure complex models such as
    neural machine translation model with attention mechanism or complex memory
    connection.
Z
zhangjinchao01 已提交
45 46

-  **Efficiency**
47

G
gangliao 已提交
48 49 50 51 52 53
    In order to unleash the power of heterogeneous computing resource,
    optimization occurs at different levels of PaddlePaddle, including
    computing, memory, architecture and communication. The following are some
    examples:

      - Optimized math operations through SSE/AVX intrinsics, BLAS libraries
L
Luo Tao 已提交
54
      (e.g. MKL, OpenBLAS, cuBLAS) or customized CPU/GPU kernels.
55
      - Optimized CNN networks through MKL-DNN library.
56
      - Highly optimized recurrent networks which can handle **variable-length**
G
gangliao 已提交
57 58 59
      sequence without padding.
      - Optimized local and distributed training for models with high dimensional
      sparse data.
Z
zhangjinchao01 已提交
60 61 62

- **Scalability**

G
gangliao 已提交
63 64 65
    With PaddlePaddle, it is easy to use many CPUs/GPUs and machines to speed
    up your training. PaddlePaddle can achieve high throughput and performance
    via optimized communication.
Z
zhangjinchao01 已提交
66 67 68

- **Connected to Products**

G
gangliao 已提交
69
    In addition, PaddlePaddle is also designed to be easily deployable. At Baidu,
70
    PaddlePaddle has been deployed into products and services with a vast number
G
gangliao 已提交
71 72 73
    of users, including ad click-through rate (CTR) prediction, large-scale image
    classification, optical character recognition(OCR), search ranking, computer
    virus detection, recommendation, etc. It is widely utilized in products at
74 75
    Baidu and it has achieved a significant impact. We hope you can also explore
    the capability of PaddlePaddle to make an impact on your product.
Z
zhangjinchao01 已提交
76 77

## Installation
Y
Yi Wang 已提交
78

S
shanyi15 已提交
79
It is recommended to read [this doc](http://www.paddlepaddle.org/documentation/docs/en/1.4/beginners_guide/index_en.html) on our website.
80

Z
zhangjinchao01 已提交
81
## Documentation
82

S
shanyi15 已提交
83 84
We provide [English](http://www.paddlepaddle.org/documentation/docs/en/1.4/beginners_guide/index_en.html) and
[Chinese](http://www.paddlepaddle.org/documentation/docs/zh/1.4/beginners_guide/install/index_cn.html) documentation.
Y
Yi Wang 已提交
85

S
Shan Yi 已提交
86
- [Deep Learning 101](https://github.com/PaddlePaddle/book)
Y
Yi Wang 已提交
87

88
  You might want to start from this online interactive book that can run in a Jupyter Notebook.
Y
Yi Wang 已提交
89

S
shanyi15 已提交
90
- [Distributed Training](http://paddlepaddle.org/documentation/docs/en/1.4/user_guides/howto/training/multi_node_en.html)
Y
Yi Wang 已提交
91 92 93

  You can run distributed training jobs on MPI clusters.

S
shanyi15 已提交
94
- [Python API](http://paddlepaddle.org/documentation/docs/en/1.4/api/index_en.html)
95

Y
Yi Wang 已提交
96
   Our new API enables much shorter programs.
97

S
shanyi15 已提交
98
- [How to Contribute](http://paddlepaddle.org/documentation/docs/en/1.4/advanced_usage/development/contribute_to_paddle/index_en.html)
Z
zhangjinchao01 已提交
99

Y
Yi Wang 已提交
100
   We appreciate your contributions!
101

102
## Communication
103

104 105 106
- [Github Issues](https://github.com/PaddlePaddle/Paddle/issues): bug reports, feature requests, install issues, usage issues, etc.
- QQ discussion group: 432676488 (PaddlePaddle).
- [Forums](http://ai.baidu.com/forum/topic/list/168?pageNo=1): discuss implementations, research, etc.
Z
zhangjinchao01 已提交
107 108 109

## Copyright and License
PaddlePaddle is provided under the [Apache-2.0 license](LICENSE).