add tutorials. (#117)

* add tutorials.
* Add links of tutorial ipynbs to docs.

Author: yecol

.github/workflows/ci.yml

@@ -11,6 +11,7 @@ on:
       - '**.rst'
       - 'docs/**'
       - 'scripts/**'
+      - 'tutorials/**'
   pull_request:
     branches:
       - main
@@ -20,6 +21,7 @@ on:
       - '**.rst'
       - 'docs/**'
       - 'scripts/**'
+      - 'tutorials/**'
 
 jobs:
   build-gae:

README-zh.md

@@ -8,17 +8,20 @@
 [![Docs-en](https://shields.io/badge/Docs-English-blue?logo=Read%20The%20Docs)](https://graphscope.io/docs)
 [![Docs-zh](https://shields.io/badge/Docs-%E4%B8%AD%E6%96%87-blue?logo=Read%20The%20Docs)](https://graphscope.io/docs/zh/)
 [![Translation](https://shields.io/badge/README-English-blue)](https://github.com/alibaba/GraphScope)
+[![Playground](https://shields.io/badge/JupyterLab-Try%20GraphScope%20Now!-F37626?logo=jupyter)](https://try.graphscope.app)
 
 GraphScope 是阿里巴巴达摩院智能计算实验室研发并开源的一站式图计算平台。依托于阿里海量数据和丰富场景，与达摩院的高水平研究，GraphScope 致力于针对实际生产场景中图计算的挑战，提供一站式高效的解决方案。
 
 GraphScope 提供了 Python 客户端，能十分方便的对接上下游工作流，具有一站式、开发便捷、性能极致等特点。它整合了智能计算实验室多个重要的创新性技术，包括 [GRAPE](https://github.com/alibaba/libgrape-lite)，[MaxGraph](interactive_engine/)，[Graph-Learn](https://github.com/alibaba/graph-learn)，支持了图分析、图的交互式查询和图学习。其中核心优势包括在业界首次支持了 Gremlin 分布式编译优化，支持了算法的自动并行化、提供了企业级场景下的极致性能等。在阿里巴巴内外部应用中，GraphScope 已经证明在多个关键互联网领域（如风控，电商推荐，广告，网络安全，知识图谱等）实现了重要的业务新价值。
 
 GraphScope 整合了达摩院的多项学术研究成果，其中的核心技术曾获得数据库领域顶级学术会议 SIGMOD2017 最佳论文奖、VLDB2017 最佳演示奖、VLDB2020 最佳论文奖亚军、世界人工智能创新大赛SAIL奖等。GraphScope 的交互查询引擎的论文已被 NSDI 2021录用。还有其它围绕 GraphScope 的十多项研究成果发表在领域顶级的学术会议或期刊上，如 TODS、SIGMOD、VLDB、KDD等。
 
-欢迎阅读[白皮书](https://github.com/alibaba/GraphScope/blob/main/docs/GraphScope_whitepaper.pdf)了解更多信息。
+欢迎访问我们的网站 [graphscope.io](https://graphscope.io) 了解更多信息。
 
 ## 快速开始
 
+我们提供了一个基于 JupyterLab 的 [Playground](https://try.graphscope.app)，您可以从浏览器中在线试用 GraphScope。 
+
 GraphScope 设计在 [Kubernetes (k8s)](https://kubernetes.io/) 管理的群集上运行。为了快速上手，我们可以按照本文档的以下步骤部署一个本地 Kubernetes 集群，并加载预编译好的镜像。。
 
 ### 环境准备

README.md

@@ -10,14 +10,16 @@
 [![Docs-en](https://shields.io/badge/Docs-English-blue?logo=Read%20The%20Docs)](https://graphscope.io/docs)
 [![Docs-zh](https://shields.io/badge/Docs-%E4%B8%AD%E6%96%87-blue?logo=Read%20The%20Docs)](https://graphscope.io/docs/zh/)
 [![README-zh](https://shields.io/badge/README-%E4%B8%AD%E6%96%87-blue)](README-zh.md)
+[![Playground](https://shields.io/badge/JupyterLab-Try%20GraphScope%20Now!-F37626?logo=jupyter)](https://try.graphscope.app)
 
 GraphScope is a unified distributed graph computing platform that provides a one-stop environment for performing diverse graph operations on a cluster of computers through a user-friendly Python interface. GraphScope makes multi-staged processing of large-scale graph data on compute clusters simple by combining several important pieces of Alibaba technology: including [GRAPE](https://github.com/alibaba/libgrape-lite), [MaxGraph](interactive_engine/), and [Graph-Learn](https://github.com/alibaba/graph-learn) (GL) for analytics, interactive, and graph neural networks (GNN) computation, respectively, and the [vineyard](https://github.com/alibaba/libvineyard) store that offers efficient in-memory data transfers.
 
-Read more in the [whitepaper](https://github.com/alibaba/GraphScope/blob/main/docs/GraphScope_whitepaper.pdf).
-
+Visit our website at [graphscope.io](https://graphscope.io) to learn more.
 
 ## Getting Started
 
+We provide a [Playground](https://try.graphscope.app) with a managed JupyterLab. [Try GraphScope](https://try.graphscope.app) straight away in your browser!
+
 GraphScope can run on clusters managed by [Kubernetes](https://kubernetes.io/) within containers. For quickly getting started, we can set up a *local* Kubernetes cluster and take advantage of pre-built Docker images as follows.
 
 ### Prerequisites

docs/index.rst

@@ -22,6 +22,7 @@ and the vineyard store that offers efficient in-memory data transfers.
 
    installation
    getting_started
+   tutorials
    deployment
    loading_graph
    interactive_engine

docs/tutorials.rst

@@ -0,0 +1,25 @@
+Tutorials
+=========
+
+We provide a set of comprehensive tutorials for users to get started with GraphScope. In the tutorials you could learn
+how to launch a session, to load graphs, and to run analytical analysis, interactive queries as well as graph neural
+network trainings using GAE, GIE and GLE in GraphScope.
+
+The tutorials are organized as jupyter notebooks. Users can try them in any jupyter notebook servers. You could play
+with those tutorials using our well-prepared jupyterlab environment in `GraphScope Playground <https://try.graphscope.app>`_.
+
+All tutorials are listed as follows:
+
+.. toctree::
+   :numbered:
+   :maxdepth: 2
+   :caption: Contents
+
+   How to Create a Session <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/1_how_to_launch_a_session.ipynb>
+   Loading Graphs <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/2_loading_graphs.ipynb>
+   Built-in Analytical Algorithms <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/3_builtin_analytical_algorithms.ipynb>
+   Writing Your Own Algorithms <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/4_writing_your_own_algorithms.ipynb>
+   Interactive Query with Gremlin <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/5_interactive_query_with_gremlin.ipynb>
+   Unsupervised Learning with GraphSage <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/6_unsupervised_learning_with_graphsage.ipynb>
+   Supervised Learning with GCN <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/7_supervised_learning_with_gcn.ipynb>
+   A Complex Workflow: Node Classification on Citation Network <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/8_node_classification_on_citation_network.ipynb>

docs/zh/index.rst

@@ -33,6 +33,7 @@ GraphScope 的交互查询引擎的论文已被 NSDI 2021录用。
    installation
    getting_started
    deployment
+   tutorials
    loading_graph
    interactive_engine
    analytics_engine

docs/zh/tutorials.rst

@@ -0,0 +1,24 @@
+入门教程
+=========
+
+我们为GraphScope的用户准备了一系列详尽的入门教程。从这些入门教程中你可以了解到如何启动一个GraphScope会话，如何载图，以及如何分别
+使用GraphScope中的GAE、GIE、GLE引擎来完成图分析任务、交互式查询和图神经网络模型的训练。
+
+我们用jupyter notebook来组织这些教程，用户可以在任何一个jupyter notebook服务中尝试执行。我们还提供了已经准备好的Jupyter Lab
+环境 `GraphScope Playground <https://try.graphscope.app>`_ 以便于用户更快地上手GraphScope。
+
+入门教程包括如下这些内容：
+
+.. toctree::
+   :numbered:
+   :maxdepth: 2
+   :caption: 目录
+
+   如何创建会话 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/1_how_to_launch_a_session.ipynb>
+   载图 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/2_loading_graphs.ipynb>
+   内置图分析算法 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/3_builtin_analytical_algorithms.ipynb>
+   编写自定义图分析算法 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/4_writing_your_own_algorithms.ipynb>
+   基于Gremlin的交互式查询 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/5_interactive_query_with_gremlin.ipynb>
+   基于GraphSage的无监督学习 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/6_unsupervised_learning_with_graphsage.ipynb>
+   基于GCN的有监督学习 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/7_supervised_learning_with_gcn.ipynb>
+   论文引用网络中的节点分类任务 <https://nbviewer.jupyter.org/github/alibaba/GraphScope/blob/main/tutorials/zh/8_node_classification_on_citation_network.ipynb>

python/.gitignore

@@ -3,6 +3,7 @@ __pycache__/
 *.pyc
 .ipynb_checkpoints
 *.egg-info/
+*.egg/
 build/
 dist/
 

python/graphscope/client/session.py

@@ -71,7 +71,9 @@
         os.path.join(__file__, "..", "..", "..", "..", "coordinator")
     )
 
-DEFAULT_CONFIG_FILE = os.environ.get("GS_CONFIG_PATH", "")
+DEFAULT_CONFIG_FILE = os.environ.get(
+    "GS_CONFIG_PATH", os.path.expanduser("~/.graphscope/session.json")
+)
 
 _session_dict = {}
 
@@ -330,7 +332,7 @@ def __init__(
         if isinstance(config, dict):
             self._config_params.update(config)
         elif isinstance(config, str):
-            self._load_config(config)
+            self._load_config(config, False)
         elif DEFAULT_CONFIG_FILE:
             self._load_config(DEFAULT_CONFIG_FILE)
 
@@ -432,11 +434,15 @@ def __str__(self):
     def session_id(self):
         return self._session_id
 
-    def _load_config(self, path):
+    def _load_config(self, path, slient=True):
         config_path = os.path.expandvars(os.path.expanduser(path))
-        with open(config_path, "r") as f:
-            data = json.load(f)
-            self._config_params.update(data)
+        try:
+            with open(config_path, "r") as f:
+                data = json.load(f)
+                self._config_params.update(data)
+        except Exception as exp:  # noqa
+            if not slient:
+                raise exp
 
     @property
     def info(self):