run-in-15-minutes.en.md

Run in 15 minutes

Install python3.x with conda

Open miniconda official website to download the miniconda installation package, modify the permission and install.

$ 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

Allow conda init the default .bashrc environment (zsh users also need to modify the conda configuration in .zshrc by self).

$ conda --version
conda 4.10.3

Create a Python3.x (take 3.8 as an example) environment and activate it.

$ conda create --name py38 python=3.8
$ conda activate py38

Install Prebuilt package

Download the .whl package corresponding to the python version from MegFlow release, install

$ python3 -m pip install megflow-0.1.0-py38-none-linux_x86_64.whl --force-reinstall

After completion, import megflow should works.

$ 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 megflow

.whl provides the megflow_run command, some environments may require export PATH=~/.local/bin/:${PATH}

$ apt install build-essential -y
$ megflow_run -h
megflow_run 1.0
megvii
...

Power On Self Test

Download MegFlow source code (we need flow-python/examples)

$ git clone https://github.com/MegEngine/MegFlow.git

Run Test

$ cd ${MegFlow_PATH}/flow-python/examples
$ megflow_run -p logical_test

logical_test is the most basic calculation graph test case. This operation indicates that the MegFlow compilation is successful and there is no problem with the basic semantics.

megflow_run is the implementation of computational graphs. Do not need to care about Rust/cargo, just make sure that

• import megflow succeed
• megflow_run -h works well

Python Built-in Applications

Next, start running Python application

• cat finder
• electric bicycle detection
• quickstart
• video super resolution