- What is Explainable AI (XAI)
- Installation
- Note
- Quick Start and API References
- Community
- Contributing
- License
XAI is an explainable AI toolbox based on MindSpore. Currently, most deep learning models are black-box models with good performance but poor explainability. XAI aims to provide users decision explanations of the model, helps users better understand the model, trust the model, and improve the model when an error occurs. Besides a variety of explanation methods, we also provide a set of evaluation methods to evaluate the explanation methods from various dimensions. It helps users compare and select the explanation methods that are most suitable for a particular scenario.
- OS: EulerOS-aarch64, CentOS-aarch64, CentOS-x86, Ubuntu-aarch64 or Ubuntu-x86
- Device: Ascend 910 or GPU CUDA 10.1, 11.1
- Python 3.7.5 or 3.9.0
- MindSpore 1.7 or above
The simplest way to do it:
pip install mindspore-xai
Or you may install with a downloaded .whl
package from MindSpore XAI download page:
pip install mindspore_xai-{version}-py3-none-any.whl
- Download source code from gitee.com:
git clone https://gitee.com/mindspore/xai.git
- Install the dependency python modules:
cd xai
pip install -r requirements.txt
- Install the XAI module from source code:
python setup.py install
- Optionally, you may build a
.whl
package for installation without step 3:
bash package.sh
pip install output/mindspore_xai-{version}-py3-none-any.whl
Upon successful installation, importing 'mindspore_xai' module in python will cause no error:
import mindspore_xai
print(mindspore_xai.__version__)
For the quick start guide and API references, please checkout XAI Homepage.
Checkout how MindSpore Open Governance works
Welcome contributions. See our Contributor Wiki for more details.