Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document.
Note: Even the most advanced machine translation cannot match the quality of professional translators.
Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
Intended Audience
You need to develop custom TBE operators in the following situations:
- The Ascend AI software stack does not support operators on a specific network.
- You would like to develop your own operators to improve computing performance.
- You would like to modify the computation logic of an operator.
This document is intended for developers who develop operators using the TBE. After reading this document, you will be able to:
- Understand the implementation principle and method of the TBE operator.
- Develop custom operators based on the samples provided in this document.
To better understand this document, you should have:
- Proficiency in Python/C++/C development
- Knowledge of mathematical expressions
- Knowledge of machine learning and deep learning
- Knowledge of TVM and the open source framework TensorFlow/Caffe