Paddlepaddle is a domestic open source framework launched by Baidu. With the blessing of Baidu's strong research and development capabilities, paddlepaddle is developing rapidly and becoming increasingly mature. At present, paddle can be used as a deep learning framework instead of python and tensorflow. At the same time, the Baidu R&D team also launched a variety of development kits, such as PaddleClas, PaddleDection, PaddleSeg and PaddleGAN. These kits can help developers quickly build their own applications.
From the beginning of this article, I will learn paddlepaddle deep learning framework with you, and build magical artificial intelligence applications together.
First we learn how to create a paddlepaddle development environment. The latest version of paddlepaddle is 2.0.0rc1, so we install the latest version of the framework today.
CPU version:
Take mac os system as an example, you can use pip to install, execute the following command:
python -m pip install paddlepaddle==2.0.0rc1 -i https://mirror.baidu.com/pypi/simple
Linux and Windows system versions are similar.
GPU version:
First, you need to install the cuda library, and different versions of CUDA can be installed according to the actual situation.
Different versions of CUDA environment have different installation commands.
MIRACLES 9.0
python -m pip install paddlepaddle-gpu==2.0.0rc1.post90 -f https://paddlepaddle.org.cn/whl/stable.html
MIRACLES 10.0
python -m pip install paddlepaddle-gpu==2.0.0rc1.post100 -f https://paddlepaddle.org.cn/whl/stable.html
MIRACLES 10.1
python -m pip install paddlepaddle-gpu==2.0.0rc1.post101 -f https://paddlepaddle.org.cn/whl/stable.html
MIRACLES 10.2
python -m pip install paddlepaddle-gpu==2.0.0rc1 -i https://mirror.baidu.com/pypi/simple
MIRACLES 11.0
python -m pip install paddlepaddle-gpu==2.0.0rc1.post110 -f https://paddlepaddle.org.cn/whl/stable.html
At the same time, if you use Baidu's AI Studio, you can omit the above installation steps. AI Studion has a built-in PaddlePaddle development environment and free 10 hours of Nvidia V100 computing power every day.
The address of AI Studio is as follows:
https://aistudio.baidu.com/aistudio/index
After the installation is complete, you can use the following methods to verify whether the environment is installed successfully.
python
>>> import paddle
>>> paddle.fluid.install_check.run_check()
If the following content is output, the installation is successful.
Your Paddle Fluid is installed successfully! Let's start deep Learning with Paddle Fluid now