309 Star 1.3K Fork 557

MindSpore / docs

Create your Gitee Account
Explore and code with more than 8 million developers,Free private repositories !:)
Sign up
Clone or Download
mindspore_ascend_install_pip_en.md 13.64 KB
Copy Edit Web IDE Raw Blame History
yanghaoran authored 2022-05-12 21:21 . add Ascend packages info for 1.7.0

Installing MindSpore in Ascend 910 by pip

This document describes how to quickly install MindSpore in a Linux system with an Ascend 910 environment by pip.

  • If you want to install MindSpore by pip on an EulerOS 2.8 with the configured Ascend AI processor software package, you may use automatic installation script for one-click installation, see Automatic Installation section. The automatic installation script will install MindSpore and its required dependencies.

  • If your system is one of Ubuntu 18.04/CentOS 7.6/OpenEuler 20.03/KylinV10 SP1, or some dependencies, such as Python and GCC, have been installed in your system, it is recommended to install manually by referring to the installation steps in the Manual Installation section.

Automatic Installation

Before running the automatic installation script, you need to make sure that the Ascend AI processor software package is correctly installed on your system. If it is not installed, please refer to the section Installing Ascend AI processor software package to install it.

Run the following command to obtain and run the automatic installation script. The automatic installation script only supports the installation of MindSpore>=1.6.0.

wget https://gitee.com/mindspore/mindspore/raw/master/scripts/install/euleros-ascend-pip.sh
# install MindSpore 1.7.0 and Python 3.7
# the default value of LOCAL_ASCEND is /usr/local/Ascend
MINDSPORE_VERSION=1.7.0 bash -i ./euleros-ascend-pip.sh
# to specify Python and MindSpore version, taking Python 3.9 and MindSpore 1.6.0 as examples
# and set LOCAL_ASCEND to /home/xxx/Ascend, use the following manners
# LOCAL_ASCEND=/home/xxx/Ascend PYTHON_VERSION=3.9 MINDSPORE_VERSION=1.6.0 bash -i ./euleros-ascend-pip.sh

This script performs the following operations:

  • Install the dependencies required by MindSpore, such as GCC and gmp.
  • Install Python3 and pip3 and set them as default.
  • Install MindSpore Ascend by pip.
  • Install Open MPI if OPENMPI is set to on.

After the script is executed, you need to reopen the terminal window to make the environment variables take effect.

The automatic installation script creates a virtual environment named mindspore_pyXX for MindSpore. Where XX is the Python version, such as Python 3.7, the virtual environment name is mindspore_py37. Run the following command to show all virtual environments.

conda env list

To activate the virtual environment, take Python 3.7 as an example, execute the following command.

conda activate mindspore_py37

Now you can jump to the Configuring Environment Variables section to set the relevant environment variables.

For more usage, see the script header description.

Manual Installation

The following table lists the system environment and third-party dependencies required for installing MindSpore.

software version description
Ubuntu 18.04/CentOS 7.6/EulerOS 2.8/OpenEuler 20.03/KylinV10 SP1 - OS for running MindSpore
Python 3.7-3.9 Python environment that MindSpore depends on
Ascend AI processor software package - Ascend platform AI computing library used by MindSpore
GCC 7.3.0 C++ compiler for compiling MindSpore
gmp 6.1.2 Multiple precision arithmetic library used by MindSpore
Open MPI 4.0.3 high performance message passing library used by MindSpore (optional, required for single-node/multi-GPU and multi-node/multi-GPU training)

The following describes how to install the third-party dependencies.

Installing Python

Python can be installed by Conda.

Install Miniconda:

cd /tmp
curl -O https://mirrors.tuna.tsinghua.edu.cn/anaconda/miniconda/Miniconda3-py37_4.10.3-Linux-x86_64.sh
bash Miniconda3-py37_4.10.3-Linux-x86_64.sh -b
cd -
. ~/miniconda3/etc/profile.d/conda.sh
conda init bash

After the installation is complete, you can set up Tsinghua source acceleration download for Conda, and see here.

Create a virtual environment, taking Python 3.7.5 as an example:

conda create -n mindspore_py37 python=3.7.5 -y
conda activate mindspore_py37

Run the following command to check the Python version.

python --version

If you are using an ARM architecture system, please ensure that pip installed for current Python has a version >= 19.3. If not, upgrade pip with the following command.

python -m pip install -U pip

Installing Ascend AI processor software package

For detailed installation guide, please refer to Ascend Data Center Solution 22.0.RC1 Installation Guide.

The default installation path of the installation package is /usr/local/Ascend. Ensure that the current user has the right to access the installation path of Ascend AI processor software package, If not, the root user needs to add the current user to the user group where /usr/local/Ascend is located.

Install the .whl packages provided in Ascend AI processor software package. The .whl packages are released with the software package. If the .whl packages have been installed before, you need to uninstall the packages by the following command.

pip uninstall te topi hccl -y

Run the following command to install the .whl packages if the Ascend AI package has been installed in default path. If the installation path is not the default path, you need to replace the path in the command with the installation path.

pip install /usr/local/Ascend/ascend-toolkit/latest/fwkacllib/lib64/topi-*-py3-none-any.whl
pip install /usr/local/Ascend/ascend-toolkit/latest/fwkacllib/lib64/te-*-py3-none-any.whl
pip install /usr/local/Ascend/ascend-toolkit/latest/fwkacllib/lib64/hccl-*-py3-none-any.whl

Installing GCC

  • On Ubuntu 18.04, run the following commands to install.

    sudo apt-get install gcc-7 -y
  • On CentOS 7, run the following commands to install.

    sudo yum install centos-release-scl
    sudo yum install devtoolset-7

    After installation, run the following commands to switch to GCC 7.

    scl enable devtoolset-7 bash
  • On EulerOS and OpenEuler, run the following commands to install.

    sudo yum install gcc -y

Installing gmp

  • On Ubuntu 18.04, run the following commands to install.

    sudo apt-get install libgmp-dev -y
  • On CentOS 7, EulerOS and OpenEuler, run the following commands to install.

    sudo yum install gmp-devel -y

Installing Open MPI-optional

Run the following command to compile and install Open MPI.

curl -O https://download.open-mpi.org/release/open-mpi/v4.0/openmpi-4.0.3.tar.gz
tar xzf openmpi-4.0.3.tar.gz
cd openmpi-4.0.3
./configure --prefix=/usr/local/openmpi-4.0.3
sudo make install
echo -e "export PATH=/usr/local/openmpi-4.0.3/bin:\$PATH" >> ~/.bashrc
echo -e "export LD_LIBRARY_PATH=/usr/local/openmpi-4.0.3/lib:\$LD_LIBRARY_PATH" >> ~/.bashrc
source ~/.bashrc
cd -

Installing MindSpore

First, refer to Version List to select the version of MindSpore you want to install, and perform SHA-256 integrity check. Taking version 1.7.0 as an example, execute the following commands.

export MS_VERSION=1.7.0

Then run the following commands to install MindSpore according to the system architecture and Python version.

# x86_64 + Python3.7
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/x86_64/mindspore_ascend-${MS_VERSION/-/}-cp37-cp37m-linux_x86_64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple
# x86_64 + Python3.8
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/x86_64/mindspore_ascend-${MS_VERSION/-/}-cp38-cp38-linux_x86_64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple
# x86_64 + Python3.9
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/x86_64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_x86_64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple
# aarch64 + Python3.7
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp37-cp37m-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple
# aarch64 + Python3.8
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp38-cp38-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple
# aarch64 + Python3.9
pip install https://ms-release.obs.cn-north-4.myhuaweicloud.com/${MS_VERSION}/MindSpore/ascend/aarch64/mindspore_ascend-${MS_VERSION/-/}-cp39-cp39-linux_aarch64.whl --trusted-host ms-release.obs.cn-north-4.myhuaweicloud.com -i https://pypi.tuna.tsinghua.edu.cn/simple

When the network is connected, dependencies of MindSpore are automatically downloaded during the .whl package installation. (For details about the dependency, see required_package in setup.py .) In other cases, you need to install it by yourself. When running models, you need to install additional dependencies based on requirements.txt specified for different models in ModelZoo. For details about common dependencies, see requirements.txt.

Configuring Environment Variables

If Ascend AI processor software is installed in a non-default path, after MindSpore is installed, export runtime-related environment variables. /usr/local/Ascend in the following command LOCAL_ASCEND=/usr/local/Ascend denotes the installation path of the software package, and you need to replace it as the actual installation path of the software package.

# control log level. 0-DEBUG, 1-INFO, 2-WARNING, 3-ERROR, 4-CRITICAL, default level is WARNING.
export GLOG_v=2

# Conda environmental options
LOCAL_ASCEND=/usr/local/Ascend # the root directory of run package

# lib libraries that the run package depends on
export LD_LIBRARY_PATH=${LOCAL_ASCEND}/ascend-toolkit/latest/fwkacllib/lib64:${LOCAL_ASCEND}/driver/lib64:${LOCAL_ASCEND}/ascend-toolkit/latest/opp/op_impl/built-in/ai_core/tbe/op_tiling:${LD_LIBRARY_PATH}

# Environment variables that must be configured
## TBE operator implementation tool path
export TBE_IMPL_PATH=${LOCAL_ASCEND}/ascend-toolkit/latest/opp/op_impl/built-in/ai_core/tbe
## OPP path
export ASCEND_OPP_PATH=${LOCAL_ASCEND}/ascend-toolkit/latest/opp
## AICPU path
## TBE operator compilation tool path
export PATH=${LOCAL_ASCEND}/ascend-toolkit/latest/fwkacllib/ccec_compiler/bin/:${PATH}
## Python library that TBE implementation depends on

Installation Verification


python -c "import mindspore;mindspore.run_check()"

The outputs should be the same as:

MindSpore version: __version__
The result of multiplication calculation is correct, MindSpore has been installed successfully!

It means MindSpore has been installed successfully.


import numpy as np
from mindspore import Tensor
import mindspore.ops as ops
from mindspore import set_context

x = Tensor(np.ones([1,3,3,4]).astype(np.float32))
y = Tensor(np.ones([1,3,3,4]).astype(np.float32))
print(ops.add(x, y))

The outputs should be the same as:

[[[[2. 2. 2. 2.]
   [2. 2. 2. 2.]
   [2. 2. 2. 2.]]

  [[2. 2. 2. 2.]
   [2. 2. 2. 2.]
   [2. 2. 2. 2.]]

  [[2. 2. 2. 2.]
   [2. 2. 2. 2.]
   [2. 2. 2. 2.]]]]

It means MindSpore has been installed successfully.

Version Update

Use the following command if you need to update the MindSpore version:

pip install --upgrade mindspore-ascend=={version}

Of which,

  • When updating to a release candidate (rc) version, {version} should be specified manually as the rc version number, e.g. 1.6.0rc1; When updating to a standard release, =={version} could be removed.