Benchmarking Python Tools for Automatic Differentiation

TL;DR

This paper evaluates and compares the performance and precision of various Python tools for automatic differentiation using a cluster geometry optimization problem, highlighting PyCppAD's superior performance.

Contribution

It provides a systematic benchmarking of Python AD tools, revealing PyCppAD's efficiency and comparable precision to PyADOL-C for large-scale problems.

Findings

PyCppAD has the best performance among tested tools.

PyADOL-C also performs well but slightly less efficiently.

Both tools maintain high precision in gradient calculations.

Abstract

In this paper we compare several Python tools for automatic differentiation. In order to assess the difference in performance and precision, the problem of finding the optimal geometrical structure of the cluster with identical atoms is used as follows. First, we compare performance of calculating gradients for the objective function. We showed that the PyADOL-C and PyCppAD tools have much better performance for big clusters than the other ones. Second, we assess precision of these two tools by calculating the difference between the obtained at the optimal configuration gradient norms. We conclude that PyCppAD has the best performance among others, while having almost the same precision as the second- best performing tool - PyADOL-C.