Installation Manuals¶
The manuals will guide you to build and install PaddlePaddle on your 64-bit desktop or laptop.
1. Operating system requirements:¶
Windows 7 / 8 / 10, Pro/Enterprise
Ubuntu 14.04 / 16.04 / 18.04
CentOS 6 / 7
MacOS 10.11 / 10.12 / 10.13 / 10.14
64-bit operating system is required
2. Processor requirements:¶
Processor supports MKL
The processor architecture is x86_64(or called x64, Intel 64, AMD64). Currently, PaddlePaddle does not support arm64.
3. Version requirements of python and pip:¶
Python 2 requires version 2.7.15+
Python 3 requires version 3.5.1+/3.6/3.7
Python needs pip, and pip requires version 9.0.1+
Python and pip requires 64-bit
4. PaddlePaddle’s support for GPU:¶
Currently, PaddlePaddle only supports CUDA driver of NVIDIA graphics card.
You need to install cuDNN , and version 7.6+ is required(For CUDA9/10)
If you need GPU multi-card mode, you need to install NCCL 2
Only Ubuntu/CentOS support NCCL 2
You need to install CUDA , depending on your system, there are different requirements for CUDA version:
Windows install GPU version
Windows 7 / 8 / 10 support CUDA 9.0 / 10.0 single-card mode, but don’t support CUDA 9.1/9.2/10.1
don’t support install using nvidia-docker
Ubuntu install GPU version
Ubuntu 14.04 supports CUDA 10.0/10.1, but doesn’t support CUDA 9.0/9.1/9.2
Ubuntu 16.04 supports CUDA 9.0/9.1/9.2/10.0/10.1
Ubuntu 18.04 supports CUDA 10.0/10.1, but doesn’t support CUDA 9.0/9.1/9.2
If you install using nvidia-docker , it supports CUDA 9.0/9.1/9.2/10.0/10.1
CentOS install GPU version
If you install using native pip :
CentOS 7 supports CUDA 9.0/9.1/9.2/10.0/10.1, CUDA 9.1 supports single-card mode only
CentOS 6 supports CUDA 9.0/9.1/9.2/10.0/10.1 single-card mode
If you compile and install using native source code:
CentOS 7 supports CUDA 9.0/9.1/9.2/10.0/10.1, CUDA 9.1 supports single-card mode only
CentOS 6 is not recommended, we don’t provide official support in case of compilation problems
If you install using nvidia-docker , CentOS 7 supports CUDA 9.0/9.1/9.2/10.0/10.1
MacOS isn’t supported: PaddlePaddle has no GPU support in Mac OS platform
Please make sure your environment meets the above conditions. If you have other requirements, please refer to Appendix .
5. PaddlePaddle’s support for NCCL:¶
Support for Windows
not support NCCL
Support for Ubuntu
Ubuntu 14.04:
support NCCL v2.4.2-v2.4.8 under CUDA10.1
support NCCL v2.3.7-v2.4.8 under CUDA10.0
Ubuntu 16.04:
support NCCL v2.4.2-v2.4.8 under CUDA10.1
support NCCL v2.3.7-v2.4.8 under CUDA10.0/9.2/9.0
support NCCL v2.1.15 under CUDA9.1
Ubuntu 18.04:
support v2.4.2-v2.4.8 under CUDA10.1
support NCCL v2.3.7-v2.4.8 under CUDA10.0
Support for CentOS
CentOS 6: not support NCCL
CentOS 7:
support NCCL v2.4.2-v2.4.8 under CUDA10.1
support NCCL v2.3.7-v2.4.8 under CUDA10.0/9.2/9.0
Support for MacOS
not support NCCL
The first way to install: use pip to install¶
You can choose any of the four ways to install: “use pip to install”, “use Conda to install”, “use Docker to install”, “compiling from the source code”
This section describes how to use pip to install.
You need to confirm that your operating system meets the requirements listed above
You need to confirm that your processor meets the requirements listed above
Confirm that the Python where you need to install PaddlePaddle is your expected location, because your computer may have multiple Python
If you are using Python 2, use the following command to output Python path. Depending on your environment, you may need to replace Python in all command lines in the description with specific Python path
In the Windows environment, the command to output Python path is:
where python
In the MacOS/Linux environment, the command to output Python path is:
which python
If you are using Python 3, use the following command to output Python path. Depending on your environment, you may need to replace Python in all command lines in the description with specific Python path
In the Windows environment, the command to output Python path is:
where python3
In the MacOS/Linux environment, the command to output Python path is:
which python3
Check the version of Python
If you are using Python 2，confirm it is 2.7.15+ using command
python --version
If you are using Python 3，confirm it is 3.5.1+/3.6/3.7 using command
python3 --version
Check the version of pip and confirm it is 9.0.1+
If you are using Python 2
python -m ensurepip python -m pip --version
If you are using Python 3
python3 -m ensurepip python3 -m pip --version
Confirm that Python and pip is 64 bit，and the processor architecture is x86_64(or called x64、Intel 64、AMD64)architecture. Currently, PaddlePaddle doesn’t support arm64 architecture. The first line below outputs “64bit”, and the second line outputs “x86_64”, “x64” or “AMD64” :
If you use Python 2
python -c "import platform;print(platform.architecture()[0]);print(platform.machine())"
If you use Python 3
python3 -c "import platform;print(platform.architecture()[0]);print(platform.machine())"
If you want to use pip to install PaddlePaddle, you can use the command below directly:
(1). CPU version : If you only want to install CPU version, please refer to command below
If you are using Python 2, command to install CPU version is:
python -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple or python -m pip install paddlepaddle -i https://pypi.tuna.tsinghua.edu.cn/simple
If you are using Python 3, command to install CPU version is:
python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple or python3 -m pip install paddlepaddle -i https://pypi.tuna.tsinghua.edu.cn/simple
(2). GPU version : If you only want to install GPU version, please refer to command below
Note:
You need to confirm that your GPU meets the requirements listed above
If you are using Python2, please attention that PaddlePaddle installed through command below supports CUDA10.0 under Windows、Ubuntu、CentOS by default:
python -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple or python -m pip install paddlepaddle-gpu -i https://pypi.tuna.tsinghua.edu.cn/simple
If you are using Python 2, CUDA 9, cuDNN 7.3+, command to install GPU version:
python -m pip install paddlepaddle-gpu==1.8.4.post97 -i https://mirror.baidu.com/pypi/simple or python -m pip install paddlepaddle-gpu==1.8.4.post97 -i https://pypi.tuna.tsinghua.edu.cn/simple
If you are using Python 2, CUDA 10.0, cuDNN 7.3+, command to install GPU version:
python -m pip install paddlepaddle-gpu==1.8.4.post107 -i https://mirror.baidu.com/pypi/simple or python -m pip install paddlepaddle-gpu==1.8.4.post107 -i https://pypi.tuna.tsinghua.edu.cn/simple
If you are using Python 3, please change python in the above command to python3 and install.
Verify installation
After the installation is complete, you can use python or python3 to enter the Python interpreter and then use import paddle.fluid as fluid and then fluid.install_check.run_check() to verify that the installation was successful.
If Your Paddle Fluid is installed succesfully! appears, it means the installation was successful.
For more information to help, please refer to:
The second way to install: use Conda to install¶
You can choose any of the four ways to install: “use pip to install”, “use Conda to install”, “use Docker to install”, “compiling from the source code”
This section describes how to use Conda to install.
You need to confirm that your operating system meets the requirements listed above
You need to confirm that your processor meets the requirements listed above
Confirm that the Python where you need to install PaddlePaddle is your expected location, because your computer may have multiple Python
For domestic users unable to connect to the official source of anaconda, you can add Tsinghua source for installation according to the following command.
conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/ conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/ conda config --add channels https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Paddle/ conda config --set show_channel_urls yes
If you need to create a new Conda virtual environment specifically for Paddle to use (the environment name after –name, you can choose by yourself):
If you are using Python2 under Window
conda create --name paddle python=2.7 activate paddle
If you are using Python2 under MacOS/Linux
conda create --name paddle python=2.7 conda activate paddle
If you are using Python3 under Window, note: python3 version can be 3.5.1+/3.6/3.7
conda create --name paddle python=3.7 activate paddle
If you are using Python3 under MacOS/Linux, note: python3 version can be 3.5.1+/3.6/3.7
conda create --name paddle python=3.7 conda activate paddle
Confirm that the Python where you need to install PaddlePaddle is your expected location, because your computer may have multiple Python, enter Anaconda’s command-line terminal, enter the following instructions to confirm the Python location
If you use Python 2, use the following command to output Python path. Depending on your environment, you may need to replace Python in all command lines in the description with specific Python path
In the Windows environment, the command to output Python path is:
where python
In the MacOS/Linux environment, the command to output Python path is:
which python
If you use Python 3, use the following command to output Python path. Depending on your environment, you may need to replace Python3 in all command lines in the description with Python or specific Python path
In the Windows environment, the command to output Python path is:
where python3
In the MacOS/Linux environment, the command to output Python path is:
which python3
Check the version of Python
If you are using Python 2, use the following command to confirm it is 2.7.15+
python --version
If you are using Python 3, use the following command to confirm it is 3.5.1+/3.6/3.7
python3 --version
Check the version of pip and confirm it is 9.0.1+
If you are using Python 2
python -m ensurepip python -m pip --version
If you are using Python 3
python3 -m ensurepip python3 -m pip --version
Confirm Python and pip is 64 bit, and the processor architecture is x86_64(or called x64,Intel 64,AMD64) architecture. Currently, PaddlePaddle doesn’t support arm64 architecture. The first line below outputs “64bit”, and the second line outputs “x86_64”, “x64” or “AMD64”:
If you are using Python 2
python -c "import platform;print(platform.architecture()[0]);print(platform.machine())"
If you are using Python 3
python3 -c "import platform;print(platform.architecture()[0]);print(platform.machine())"
If you want to use Conda to install PaddlePaddle, you can directly use commands below:
(1). CPU version :If you just want to install the CPU version, please refer to the following command installation
conda install paddlepaddle
(2). GPU version :If you just want to install the GPU version, please refer to the following command installation
Note:
You need to confirm that your GPU meets the requirements listed above
If you are using CUDA 8，cuDNN 7.1+, the command to install GPU version:
conda install paddlepaddle-gpu cudatoolkit=8.0
If you are using CUDA 9，cuDNN 7.3+, the command to install GPU version:
conda install paddlepaddle-gpu cudatoolkit=9.0
If you are using CUDA 10.0，cuDNN 7.3+, the command to install GPU version:：
conda install paddlepaddle-gpu cudatoolkit=10.0
Verify installation
After the installation is complete, you can use python or python3 to enter the Python interpreter and then use import paddle.fluid as fluid and then fluid.install_check.run_check() to verify that the installation was successful.
If Your Paddle Fluid is installed succesfully! appears, it means the installation was successful.
For more information to help, please refer to:
The third way to install: use Docker to install¶
You can choose any of the four ways to install: “use pip to install”, “use Conda to install”, “use Docker to install”, “compiling from the source code”
This section describes how to use Docker to install.
If you want to use docker to install PaddlePaddle, you can use command below:
CPU version
(1). At first you need to install docker
Note:
CentOS 6 not support docker installation
processor need supporting MKL
(2). Pull the image of the preinstalled PaddlePaddle:
docker pull hub.baidubce.com/paddlepaddle/paddle:1.8.4
(3). Use the image to build and enter the Docker container:
docker run --name paddle -it -v dir1:dir2 hub.baidubce.com/paddlepaddle/paddle:1.8.4 /bin/bash > --name [Name of container] set the name of Docker; > -it The parameter indicates that the container has been operated interactively with the local machine; > -v Parameter is used to share files between the host and the container. dir1 is the host directory and dir2 is the directory mounted inside the container. Users can customize their own mounting directory by setting dir1 and dir2.For example, $PWD:/paddle specifies to mount the current path of the host (PWD variable in Linux will expand to the absolute path of the current path) to the /paddle directory inside the container; > hub.baidubce.com/paddlepaddle/paddle:1.8.4 is the image name you need to use；/bin/bash is the command to be executed in Docker
GPU version
(1). At first you need to install nvidia-docker
Note:
processor need supporting MKL
Your computer needs to have NVIDIA graphics card supporting CUDA driver
You need to install cuDNN ，version requires 7.3+(For CUDA9/10), 7.1+(For CUDA 8)
If you need GPU multi-card mode, you need to install NCCL 2
Only Ubuntu/CentOS support NCCL 2 technology
You need to install CUDA , depending on your system, there are different requirements for CUDA version:
Ubuntu/CentOS 7 ，if you use nvidia-docker to install, CUDA 8.0/9.0/9.1/9.2/10.0 is supported
Windows/MacOS/CentOS 6 not support nvidia-docker to install
(2). Pull the image that supports CUDA 10.0, cuDNN 7.3 + pre installed PaddlePaddle:
nvidia-docker pull hub.baidubce.com/paddlepaddle/paddle:1.8.4-gpu-cuda10.0-cudnn7
(3). Use the image to build and enter the docker container:
nvidia-docker run --name paddle -it -v dir1:dir2 hub.baidubce.com/paddlepaddle/paddle:1.8.4-gpu-cuda10.0-cudnn7 /bin/bash > --name [Name of container] set name of Docker; > -it The parameter indicates that the container has been operated interactively with the local machine; > -v Parameter is used to share files between the host and the container. dir1 is the host directory and dir2 is the directory mounted inside the container. Users can customize their own mounting directory by setting dir1 and dir2.For example, $PWD:/paddle specifies to mount the current path of the host (PWD variable in Linux will expand to the absolute path of the current path) to the /paddle directory inside the container; > hub.baidubce.com/paddlepaddle/paddle:1.8.4 is the image name you need to use；/bin/bash is the command to be executed in Docker
Or if you need the version supporting CUDA 9, replace cuda10.0 of the above command with cuda9.0
If your machine is not in China’s mainland , you can pull the image directly from DockerHub:
docker run --name paddle -it -v dir1:dir2 paddlepaddle/paddle:1.8.4 /bin/bash > --name [Name of container] set name of Docker; > -it The parameter indicates that the container has been operated interactively with the local machine; > -v Parameter is used to share files between the host and the container. dir1 is the host directory and dir2 is the directory mounted inside the container. Users can customize their own mounting directory by setting dir1 and dir2.For example, $PWD:/paddle specifies to mount the current path of the host (PWD variable in Linux will expand to the absolute path of the current path) to the /paddle directory inside the container; > paddlepaddle/paddle:1.8.4 is the image name you need to use；/bin/bash is the command to be executed in docker
Verify installation
After the installation is complete, you can use python or python3 to enter the Python interpreter and then use import paddle.fluid as fluid and then fluid.install_check.run_check() to verify that the installation was successful.
If Your Paddle Fluid is installed succesfully! appears, it means the installation was successful.
For more help, refer to:
The fourth way to install: compile and install with source code¶
If you use PaddlePaddle only, we suggest you to choose one of the three installation methods pip, conda, docker to install.
If you need to develop PaddlePaddle, please refer to compile from source code