Learn how the Python API for the AUTOSAR Adaptive platform can help engineers and researchers to focus on the mathematical problems instead of dealing with the excess compilation time and possible debugging quirks of a C++ program. Analyzing the high-level APIs of the most widely used machine learning frameworks such as Tensorflow, PyTorch, Keras, Gluon, Chainer, and ONNX, it’s easy to recognize that the dominance of the Python language is overwhelming. The intensive computations running on a GPU are programmed in low-level code, but Python is convenient for defining and configuring the algorithms in high-level code.
Rapid prototyping with a Python API for AUTOSAR Adaptive
Using Python binding provides the balance between the simplicity of the Python language and the versatility of the AUTOSAR Adaptive Platform. As the most widely used language of major Machine Learning (ML) frameworks, Python helps data scientists deliver results quickly in automotive prototypes. The resemblance of the API names allows the team to quickly familiarize themselves with the AUTOSAR concepts and the syntax of the final C++ product. Recognizing the gap between the high-level APIs, the ML frameworks, and their automotive application, Siemens has created a set of Python packages for AUTOSAR. These packages provide simple, high-level interfaces between Python programs and AUTOSAR Adaptive.