TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials
TorchANI: A Free and Open Source PyTorch-Based Deep Learning Implementation of the ANI Neural Network Potentials
复制标题
DOI：
10.1021/acs.jcim.0c00451
复制
发表时间：
2020-07-27
影响因子：
5.6
通讯作者：
Roitberg, Adrian E.
中科院分区：
文献类型：
--
作者：
Gao, Xiang;Ramezanghorbani, Farhad;Roitberg, Adrian E.
本文提出了一个基于PyTorch的程序TorchANI，用于ANI(Anakin-ME)深度学习模型的训练/推理，以获得分子体系的势能面和其他物理性质。ANI是一种精确的神经网络潜力，最初是在一个名为NeuroChem的程序中使用C++/CUDA实现的。与NeuroChem相比，TorchANI的设计强调轻量级、用户友好、跨平台、易于阅读和修改，以实现快速原型设计，同时允许在运行性能上做出可接受的牺牲。由于原子环境向量和原子神经网络的计算都是使用PyTorch运算符实现的，因此TorchANI能够使用PyTorch的Autograd引擎自动计算解析力和海森矩阵，以及进行力训练，而不需要任何额外的代码。TorchANI是开源的，可以在giHub上免费获得：https://github.com/aiqm/torchani.
This paper presents TorchANI, a PyTorch-based program for training/inference of ANI (ANAKIN-ME) deep learning models to obtain potential energy surfaces and other physical properties of molecular systems. ANI is an accurate neural network potential originally implemented using C++/CUDA in a program called NeuroChem. Compared with NeuroChem, TorchANI has a design emphasis on being lightweight, user friendly, cross platform, and easy to read and modify for fast prototyping, while allowing acceptable sacrifice on running performance. Because the computation of atomic environmental vectors and atomic neural networks are all implemented using PyTorch operators, TorchANI is able to use PyTorch's autograd engine to automatically compute analytical forces and Hessian matrices, as well as do force training without requiring any additional codes. TorchANI is open-source and freely available on GitHub: https://github.com/aiqm/torchani.