Skip to content

Latest commit

 

History

History
74 lines (48 loc) · 1.91 KB

install.md

File metadata and controls

74 lines (48 loc) · 1.91 KB
Raw

简体中文 | English

Installation

Introduction

This document introduces how to install PaddlePaddle、PaddleVideo and its requirements.

Install PaddlePaddle

Python 3.7, CUDA 10.1, CUDNN7.6.4 nccl2.1.2 and later version are required at first, For now, PaddleVideo only support training on the GPU device. Please follow the instructions in the Installation if the PaddlePaddle on the device is lower than v2.0

Install PaddlePaddle

pip3 install paddlepaddle-gpu --upgrade

or compile from source code, please refer to Installation.

Verify Installation

import paddle
paddle.utils.run_check()

Check PaddlePaddle version：

python3 -c "import paddle; print(paddle.__version__)"

Note:

  • Make sure the compiled version is later than PaddlePaddle2.0.
  • Indicate WITH_DISTRIBUTE=ON when compiling, Please refer to Instruction for more details.
  • When running in the docker, in order to ensure that the container has enough shared memory for data read acceleration of Paddle, please set the parameter --shm_size=32g at creating a docker container, if conditions permit, you can set it to a larger value.

Install PaddleVideo

Clone PaddleVideo:

cd path_to_clone_PaddleVideo
git clone https://github.com/PaddlePaddle/PaddleVideo.git

Install requirements

python3.7 -m pip install --upgrade pip
pip3.7 install --upgrade -r requirements.txt

Install python package

pip3.7 install ppvideo==2.3.0

use scripts:

ppvideo --model_name='ppTSM_v2' --use_gpu=False --video_file='data/example.avi'