docs(build): add Dockerfile and prebuilt doc

Dockerfile

@@ -0,0 +1,31 @@
+FROM ubuntu:18.04
+LABEL maintainer=megengine@megvii.com
+
+RUN apt update \
+    && apt install -y curl \
+    && apt install -y ffmpeg \
+    && apt install -y yasm \
+    && apt install -y clang \
+    && apt install -y redis-server \
+    && apt install -y python3 \
+    && apt install -y python3-pip \
+    && apt install -y git \
+    && apt install -y vim \
+    && apt install -y build-essential
+
+RUN curl --proto '=https' --tlsv1.2 https://sh.rustup.rs -sSf -o run.sh \
+    && chmod a+x run.sh \
+    && ./run.sh -y \
+    && export PATH=$HOME/.cargo/bin:${PATH} \
+    && cargo --version
+
+RUN mkdir -p $HOME/megflow-runspace
+WORKDIR $HOME/megflow-runspace
+COPY . $HOME/megflow-runspace/
+
+RUN PATH=$HOME/.cargo/bin:${PATH} \
+    && cargo build \
+    && cd flow-python \
+    && python3 setup.py install --user \
+    && cd examples \
+    && cargo run --example run_with_plugins -- -p logical_test

README.md

@@ -11,20 +11,38 @@ MegFlow 是一个面向视觉应用的流式计算框架, 目标是简单、高
 - 支持 demux/reorder/transform 等通用函数式组件
 - Python 插件内置有栈协程，不依赖 asyncio
 - 基础测试工具，支持插件沙盒，用于单测插件
-  
+
 ## HowTo
-* [how to build and run in 15 minutes](docs/how-to-build.zh.md)
+* [how to run in 15 minutes](docs/how-to-run-in-15-minutes.zh.md)
+* [how to build with docker](docs/how-to-build-with-docker.zh.md)
+* [how to build from source](docs/how-to-build-from-source.zh.md)
 * [how to add my service](docs/how-to-add-graph.zh.md)
 * [how to add plugins](docs/how-to-add-plugins.zh.md)
 * [how to optimize and debug](docs/how-to-debug.zh.md)
 * [how to contribute](docs/how-to-contribute.zh.md)
 * [FAQ](docs/FAQ.zh.md)
 
+## Current Support List
+
+| 系统环境 | 支持情况 |
+| - | - |
+| Windows 10 WSL ubuntu18.04 | ✔️ |
+| x86 Ubuntu16.04 有 GPU | ✔️ |
+| x86 Ubuntu18.04 无 GPU | ✔️ |
+| ARM | / |
+
+| Python 版本 | 支持情况 |
+| - | - |
+| 3.6 | ✔️ |
+| 3.7 | ✔️ |
+| 3.8 | ✔️ |
+| 3.9 | / |
+
 ## Built-in Applications
 * 猫猫围栏
 * 电梯电动车报警
 
-## Coming soon
+## Coming Soon
 - 进程级别的节点、子图支持
 - 插件自动化测试部署
 - 调试工具，建图实时预览 profile 工具

docs/how-to-build.zh.md → docs/how-to-build-from-source.zh.md

@@ -1,19 +1,12 @@
 # Building from Source
 
-## Prerequisites
-
-### 软硬件环境
-
-| 测试通过的环境 | 备注 |
-| - | - |
-| win10 WSL ubuntu18.04 | - |
-| x86 Ubuntu16.04 服务器有 GPU | - |
-| x86 Ubuntu18.04 无 GPU | 运行时选 CPU config |
+如果仅仅是想运行应用，推荐使用 [docker](how-to-build-with-docker.zh.md) 或 [prebuilt .whl 包](how-to-run-in-15-minutes.zh.md) ，源码安装需要 Rust/ffmpeg 环境。
 
-支持主流 x86 Linux 版本，ARM 还在开发。
+## Prerequisites
 
 ### 安装 Rust
 ```bash
+$ sudo apt install yasm git build-essential ffmpeg curl
 $ curl --proto '=https' --tlsv1.2 https://sh.rustup.rs -sSf | sh
 ```
 
@@ -41,18 +34,18 @@ $ conda --version
 conda 4.10.3
 ```
 
-创建一个 Python3.8（已测试 3.6.13/3.7.11/3.8.11 可用。**3.9 暂不可用**，这里以 3.8 为例）的环境，激活。
+创建一个 Python3.x（这里以 3.8 为例） 的环境，激活。
 ```bash
 $ conda create --name py38 python=3.8
 $ conda activate py38
 ```
 
+
 ## Build
 
 MegFlow 需要编译 ffmpeg。考虑到 ffmpeg 依赖较多、本身又是常用工具，最简单的办法就是直接装 ffmpeg 把编译依赖装上
 
 ```bash
-$ sudo apt install yasm  # ffmpeg 编译依赖
 $ sudo apt install ffmpeg
 $ ffmpeg 
 ffmpeg version 3.4.8...
@@ -74,7 +67,7 @@ $ python3 setup.py install --user
 
 P.S. 默认 ffmpeg 依赖自动从 github 上拉取源码构建，这会使得首次构建的时间较长。若希望缩短首次构建时间，或者希望依赖一个指定版本的 ffmpeg，可以启用环境变量`CARGO_FEATURE_PREBUILD`并参考[rust-ffmpeg](https://github.com/zmwangx/rust-ffmpeg/wiki/Notes-on-building)自行构建
 
-## Python “开机自检”用例
+## Python“开机自检”
 ```bash
 $ cd examples
 $ cargo run --example run_with_plugins -- -p logical_test

docs/how-to-build-with-docker.zh.md

@@ -0,0 +1,33 @@
+# Building with docker
+
+## Build Docker Image
+
+MegFlow 提供了 [Dockerfile](../Dockerfile)，能够“可复现地”生成运行环境、减少依赖缺失的痛苦
+
+```bash
+$ cd MegFlow
+$ docker build -t megflow .
+```
+稍等一段时间（取决于网络和 CPU）镜像构建完成并做了基础自测
+```bash
+$ docker images
+REPOSITORY            TAG          IMAGE ID       CREATED             SIZE
+megflow               latest       c65e37e1df6c   18 hours ago        5.05GB
+```
+直接用 ${IMAGE ID} 进入开始跑应用
+```bash
+$ docker run  -p 18081:8081 -p 18082:8082  -i -t  c65e37e1df6c /bin/bash
+```
+
+## Python Built-in Applications
+
+接下来开始运行好玩的 Python 应用
+
+*  [猫猫围栏运行手册](../flow-python/examples/cat_finder/README.md)
+   *  图片注册猫猫
+   *  部署视频围栏，注册的猫离开围栏时会发通知
+   *  未注册的不会提示
+*  [电梯电瓶车告警](../flow-python/examples/electric_bicycle/README.md)
+   *  电梯里看到电瓶车立即报警
+*  Comming Soon
+   *  OCR： 通用字符识别

docs/how-to-run-in-15-minutes.zh.md

@@ -0,0 +1,79 @@
+# Run in 15 minutes
+
+
+
+## 安装 python3.x （推荐 conda）
+
+打开 [miniconda 官网](https://docs.conda.io/en/latest/miniconda.html) 下载 miniconda 安装包，修改权限并安装。
+
+```bash
+$ wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
+$ chmod a+x Miniconda3-latest-Linux-x86_64.sh
+$ ./Miniconda3-latest-Linux-x86_64.sh
+```
+
+安装时接受 conda 修改默认 .bashrc 环境变量（zsh 用户还需自行修改 .zshrc 中的 conda initialize 配置）。成功后 `conda` 可正常运行
+```bash
+$ conda --version
+conda 4.10.3
+```
+
+创建一个 Python3.x（这里以 3.8 为例） 的环境，激活。
+```bash
+$ conda create --name py38 python=3.8
+$ conda activate py38
+```
+
+## 安装 Prebuilt 包
+
+从 [MegFlow release](https://github.com/MegEngine/MegFlow/releases) 下载对应 python 版本的 .whl 包，安装
+```bash
+$  python3 -m pip install pyflow-0.1.0-py38-none-linux_x86_64.whl  --force-reinstall
+```
+完成后应该可以 `import pyflow`
+```bash
+$ python3
+Python 3.8.3 (default, May 19 2020, 18:47:26) 
+[GCC 7.3.0] :: Anaconda, Inc. on linux
+Type "help", "copyright", "credits" or "license" for more information.
+>>> import pyflow
+```
+.whl 打包了可执行文件 `run_with_plugins`，如果使用 conda 位置应该在
+```bash
+$ cd ${HOME}/miniconda3/envs/py38/lib/python3.8/site-packages/pyflow/
+$ sudo apt install build-essential -y
+$ ldd run_with_plugins         # 可看到仅依赖常见库
+$ ./run_with_plugins --help
+run_with_plugins 1.0
+megvii
+...
+```
+
+## Python“开机自检”
+
+```bash
+$ cd ${MegFlow_PATH}/flow-python/examples  # 这行必须
+$ run_with_plugins -p logical_test
+```
+
+`logical_test` 是 examples 下最基础的计算图测试用例，运行能正常结束表示 MegFlow 编译成功、基本语义无问题。
+
+此处常见问题：`error while loading shared libraries: libpython3.8.xxx`。如果使用 conda 只需要
+```bash
+$ export LD_LIBRARY_PATH=/home/`whoami`/miniconda3/pkgs/python-3.8.11-h12debd9_0_cpython/lib:${LD_LIBRARY_PATH}
+```
+
+> 工作原理：[pyflow](../flow-python/pyflow/__init__.py) 仅是一层接口，由 run_with_plugins “注入”建图/调度/优化等实现。
+
+## Python Built-in Applications
+
+接下来开始运行好玩的 Python 应用
+
+*  [猫猫围栏运行手册](../flow-python/examples/cat_finder/README.md)
+   *  图片注册猫猫
+   *  部署视频围栏，注册的猫离开围栏时会发通知
+   *  未注册的不会提示
+*  [电梯电瓶车告警](../flow-python/examples/electric_bicycle/README.md)
+   *  电梯里看到电瓶车立即报警
+*  Comming Soon
+   *  OCR： 通用字符识别

flow-python/gen_whl.sh

@@ -0,0 +1,28 @@
+# build py36~39 version
+conda activate py36
+cargo build --example run_with_plugins --release
+cp ../target/release/examples/run_with_plugins ./pyflow/
+ldd pyflow/run_with_plugins
+rm -rf ./build
+python3 whl-py36-setup.py bdist_wheel -p linux-x86_64 -d  py36_dist  --python-tag py36
+
+conda activate py37
+cargo build --example run_with_plugins --release
+ldd pyflow/run_with_plugins
+cp ../target/release/examples/run_with_plugins ./pyflow/
+rm -rf ./build
+python3 whl-py37-setup.py bdist_wheel -p linux-x86_64 -d  py37_dist  --python-tag py37
+
+conda activate py38
+cargo build --example run_with_plugins --release
+ldd pyflow/run_with_plugins
+cp ../target/release/examples/run_with_plugins ./pyflow/
+rm -rf ./build
+python3 whl-py38-setup.py bdist_wheel -p linux-x86_64 -d  py38_dist  --python-tag py38
+
+
+rm -rf dist
+mkdir dist
+cp py36_dist/pyflow-0.1.0-py36-none-linux_x86_64.whl dist/
+cp py37_dist/pyflow-0.1.0-py37-none-linux_x86_64.whl dist/
+cp py38_dist/pyflow-0.1.0-py38-none-linux_x86_64.whl dist/

flow-python/setup.py

@@ -9,11 +9,33 @@
 
 #!/usr/bin/env python
 # coding=utf-8
-from setuptools import setup
+from setuptools import setup, find_packages
 
 setup(
     name="pyflow",
     version="0.1.0",
     packages=["pyflow"],
+    author="Megvii IPU-SDK Team",
+    author_email="megengine@megvii.com",
+    url="https://github.com/MegEngine/MegFlow",
+    include_package_data=True,
+    classifiers=[
+        'Development Status :: 3 - Alpha',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: Apache Software License',
+        'Natural Language :: English',
+        'Operating System :: POSIX :: Linux',
+        'Programming Language :: C++',
+        'Programming Language :: Rust',
+        'Programming Language :: Python :: 3.8',
+        'Topic :: Software Development :: Libraries :: Application Frameworks',
+        'Topic :: Scientific/Engineering',
+        'Topic :: Scientific/Engineering :: Mathematics',
+        'Topic :: Scientific/Engineering :: Artificial Intelligence',
+        'Topic :: Software Development',
+        'Topic :: Software Development :: Libraries',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+    ],
+    python_requires='>=3.7,<3.8',
 )
 

flow-python/whl-py36-setup.py

@@ -0,0 +1,44 @@
+# -*- coding: utf-8 -*-
+# MegFlow is Licensed under the Apache License, Version 2.0 (the "License")
+#
+# Copyright (c) 2019-2021 Megvii Inc. All rights reserved.
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+
+#!/usr/bin/env python
+# coding=utf-8
+from setuptools import setup, find_packages
+
+setup(
+    name="pyflow",
+    version="0.1.0",
+    packages=["pyflow"],
+    author="Megvii IPU-SDK Team",
+    author_email="megengine@megvii.com",
+    url="https://github.com/MegEngine/MegFlow",
+    include_package_data=True,
+    classifiers=[
+        'Development Status :: 3 - Alpha',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: Apache Software License',
+        'Natural Language :: English',
+        'Operating System :: POSIX :: Linux',
+        'Programming Language :: C++',
+        'Programming Language :: Rust',
+        'Programming Language :: Python :: 3.6',
+        'Topic :: Software Development :: Libraries :: Application Frameworks',
+        'Topic :: Scientific/Engineering',
+        'Topic :: Scientific/Engineering :: Mathematics',
+        'Topic :: Scientific/Engineering :: Artificial Intelligence',
+        'Topic :: Software Development',
+        'Topic :: Software Development :: Libraries',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+    ],
+    python_requires='>=3.6,<3.7',
+    package_data={
+        "":['run_with_plugins']
+    } 
+)
+

flow-python/whl-py37-setup.py

@@ -0,0 +1,44 @@
+# -*- coding: utf-8 -*-
+# MegFlow is Licensed under the Apache License, Version 2.0 (the "License")
+#
+# Copyright (c) 2019-2021 Megvii Inc. All rights reserved.
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+
+#!/usr/bin/env python
+# coding=utf-8
+from setuptools import setup, find_packages
+
+setup(
+    name="pyflow",
+    version="0.1.0",
+    packages=["pyflow"],
+    author="Megvii IPU-SDK Team",
+    author_email="megengine@megvii.com",
+    url="https://github.com/MegEngine/MegFlow",
+    include_package_data=True,
+    classifiers=[
+        'Development Status :: 3 - Alpha',
+        'Intended Audience :: Developers',
+        'License :: OSI Approved :: Apache Software License',
+        'Natural Language :: English',
+        'Operating System :: POSIX :: Linux',
+        'Programming Language :: C++',
+        'Programming Language :: Rust',
+        'Programming Language :: Python :: 3.7',
+        'Topic :: Software Development :: Libraries :: Application Frameworks',
+        'Topic :: Scientific/Engineering',
+        'Topic :: Scientific/Engineering :: Mathematics',
+        'Topic :: Scientific/Engineering :: Artificial Intelligence',
+        'Topic :: Software Development',
+        'Topic :: Software Development :: Libraries',
+        'Topic :: Software Development :: Libraries :: Python Modules',
+    ],
+    python_requires='>=3.7,<3.8',
+    package_data={
+        "":['run_with_plugins']
+    } 
+)
+