index.rst 6.21 KB
Newer Older
Minjie Wang's avatar
Minjie Wang committed
1
2
3
4
5
.. DGL documentation master file, created by
   sphinx-quickstart on Fri Oct  5 14:18:01 2018.
   You can adapt this file completely to your liking, but it should at least
   contain the root `toctree` directive.

Minjie Wang's avatar
Minjie Wang committed
6
7
8
9
10
Overview of DGL
===============

Deep Graph Library (DGL) is a Python package built for easy implementation of
graph neural network model family, on top of existing DL frameworks (e.g.
Aston Zhang's avatar
Aston Zhang committed
11
PyTorch, MXNet, Gluon etc.).
Minjie Wang's avatar
Minjie Wang committed
12
13

DGL reduces the implementation of graph neural networks into declaring a set
VoVAllen's avatar
VoVAllen committed
14
of *functions* (or *modules* in PyTorch terminology).  In addition, DGL
Minjie Wang's avatar
Minjie Wang committed
15
16
17
18
19
20
21
22
23
24
provides:

* Versatile controls over message passing, ranging from low-level operations
  such as sending along selected edges and receiving on specific nodes, to
  high-level control such as graph-wide feature updates.
* Transparent speed optimization with automatic batching of computations and
  sparse matrix multiplication.
* Seamless integration with existing deep learning frameworks.
* Easy and friendly interfaces for node/edge feature access and graph
  structure manipulation.
Da Zheng's avatar
Da Zheng committed
25
* Good scalability to graphs with tens of millions of vertices.
Minjie Wang's avatar
Minjie Wang committed
26
27
28
29
30
31
32

To begin with, we have prototyped 10 models across various domains:
semi-supervised learning on graphs (with potentially billions of nodes/edges),
generative models on graphs, (previously) difficult-to-parallelize tree-based
models like TreeLSTM, etc. We also implement some conventional models in DGL
from a new graphical perspective yielding simplicity.

33
34
Getting Started
---------------
Minjie Wang's avatar
Minjie Wang committed
35

Minjie Wang's avatar
Minjie Wang committed
36
37
38
* :doc:`Installation<install/index>`.
* :doc:`Quickstart tutorial<tutorials/basics/1_first>` for absolute beginners.
* :doc:`User guide<guide/index>`.
39
* :doc:`用户指南(User guide)中文版<guide_cn/index>`.
Minjie Wang's avatar
Minjie Wang committed
40
41
* :doc:`API reference manual<api/python/index>`.
* :doc:`End-to-end model tutorials<tutorials/models/index>` for learning DGL by popular models on graphs.
42

Minjie Wang's avatar
Minjie Wang committed
43
..
44
  Follow the :doc:`instructions<install/index>` to install DGL.
45
  :doc:`<new-tutorial/1_introduction>` is the most common place to get started with.
Minjie Wang's avatar
Minjie Wang committed
46
  It offers a broad experience of using DGL for deep learning on graph data.
47

Minjie Wang's avatar
Minjie Wang committed
48
49
  API reference document lists more endetailed specifications of each API and GNN modules,
  a useful manual for in-depth developers.
50

Minjie Wang's avatar
Minjie Wang committed
51
  You can learn other basic concepts of DGL through the dedicated tutorials.
52

53
54
55
56
57
  * Learn constructing, saving and loading graphs with node and edge features :doc:`here<new-tutorial/2_dglgraph>`.
  * Learn performing computation on graph using message passing :doc:`here<new-tutorial/3_message_passing>`.
  * Learn link prediction with DGL :doc:`here<new-tutorial/4_link_predict>`.
  * Learn graph classification with DGL :doc:`here<new-tutorial/5_graph_classification>`.
  * Learn creating your own dataset for DGL :doc:`here<new-tutorial/6_load_data>`.
Minjie Wang's avatar
Minjie Wang committed
58
  * Learn working with heterogeneous graph data :doc:`here<tutorials/basics/5_hetero>`.
59

Minjie Wang's avatar
Minjie Wang committed
60
61
  End-to-end model tutorials are other good starting points for learning DGL and popular
  models on graphs. The model tutorials are categorized based on the way they utilize DGL APIs.
62

Minjie Wang's avatar
Minjie Wang committed
63
64
65
66
67
68
69
70
71
  * :ref:`Graph Neural Network and its variant <tutorials1-index>`: Learn how to use DGL to train
    popular **GNN models** on one input graph.
  * :ref:`Dealing with many small graphs <tutorials2-index>`: Learn how to train models for
    many graph samples such as sentence parse trees.
  * :ref:`Generative models <tutorials3-index>`: Learn how to deal with **dynamically-changing graphs**.
  * :ref:`Old (new) wines in new bottle <tutorials4-index>`: Learn how to combine DGL with tensor-based
    DGL framework in a flexible way. Explore new perspective on traditional models by graphs.
  * :ref:`Training on giant graphs <tutorials5-index>`: Learn how to train graph neural networks
    on giant graphs.
72

Minjie Wang's avatar
Minjie Wang committed
73
74
  Each tutorial is accompanied with a runnable python script and jupyter notebook that
  can be downloaded. If you would like the tutorials improved, please raise a github issue.
75
76
77
78
79
80
81

.. toctree::
   :maxdepth: 1
   :caption: Get Started
   :hidden:
   :glob:

Minjie Wang's avatar
Minjie Wang committed
82
83
   install/index
   install/backend
84
   new-tutorial/1_introduction
85

86
.. toctree::
Minjie Wang's avatar
Minjie Wang committed
87
88
   :maxdepth: 2
   :caption: User Guide
89
   :hidden:
90
   :titlesonly:
Minjie Wang's avatar
Minjie Wang committed
91
   :glob:
Minjie Wang's avatar
Minjie Wang committed
92

Minjie Wang's avatar
Minjie Wang committed
93
94
   guide/graph
   guide/message
95
   guide/nn
Minjie Wang's avatar
Minjie Wang committed
96
97
98
99
   guide/data
   guide/training
   guide/minibatch
   guide/distributed
100
   guide/mixed_precision
Minjie Wang's avatar
Minjie Wang committed
101
102
103

.. toctree::
   :maxdepth: 2
Minjie Wang's avatar
Minjie Wang committed
104
   :caption: API Reference
105
   :hidden:
Minjie Wang's avatar
Minjie Wang committed
106
   :glob:
Minjie Wang's avatar
Minjie Wang committed
107

Minjie Wang's avatar
Minjie Wang committed
108
109
   api/python/dgl
   api/python/dgl.data
110
   api/python/dgl.dataloading
111
   api/python/dgl.DGLGraph
112
   api/python/dgl.distributed
113
114
115
116
117
   api/python/dgl.function
   api/python/nn
   api/python/dgl.ops
   api/python/dgl.sampling
   api/python/udf
Minjie Wang's avatar
Minjie Wang committed
118
119
120
121
122
123
124
125

.. toctree::
   :maxdepth: 3
   :caption: Model Tutorials
   :hidden:
   :glob:

   tutorials/models/index
Minjie Wang's avatar
Minjie Wang committed
126

Minjie Wang's avatar
Minjie Wang committed
127
128
.. toctree::
   :maxdepth: 1
129
   :caption: Developer Notes
130
   :hidden:
Minjie Wang's avatar
Minjie Wang committed
131
132
   :glob:

133
   contribute
134
135
136
137
138
139
140
141
   developer/ffi

.. toctree::
   :maxdepth: 1
   :caption: Misc
   :hidden:
   :glob:

Minjie Wang's avatar
Minjie Wang committed
142
143
   faq
   env_var
144
145
   resources

Minjie Wang's avatar
Minjie Wang committed
146
147
148
149
150
151
152
Relationship of DGL to other frameworks
---------------------------------------
DGL is designed to be compatible and agnostic to the existing tensor
frameworks. It provides a backend adapter interface that allows easy porting
to other tensor-based, autograd-enabled frameworks.


153
154
155
156
157
Free software
-------------
DGL is free software; you can redistribute it and/or modify it under the terms
of the Apache License 2.0. We welcome contributions.
Join us on `GitHub <https://github.com/dmlc/dgl>`_ and check out our
158
:doc:`contribution guidelines <contribute>`.
159
160
161
162
163
164
165
166
167
168
169
170
171
172

History
-------
Prototype of DGL started in early Spring, 2018, at NYU Shanghai by Prof. `Zheng
Zhang <https://shanghai.nyu.edu/academics/faculty/directory/zheng-zhang>`_ and
Quan Gan. Serious development began when `Minjie
<https://jermainewang.github.io/>`_, `Lingfan <https://cs.nyu.edu/~lingfan/>`_
and Prof. `Jinyang Li <http://www.news.cs.nyu.edu/~jinyang/>`_ from NYU's
system group joined, flanked by a team of student volunteers at NYU Shanghai,
Fudan and other universities (Yu, Zihao, Murphy, Allen, Qipeng, Qi, Hao), as
well as early adopters at the CILVR lab (Jake Zhao). Development accelerated
when AWS MXNet Science team joined force, with Da Zheng, Alex Smola, Haibin
Lin, Chao Ma and a number of others. For full credit, see `here
<https://www.dgl.ai/ack>`_.
Minjie Wang's avatar
Minjie Wang committed
173

Minjie Wang's avatar
Minjie Wang committed
174
175
176
Index
-----
* :ref:`genindex`