Machine Learning Quantum Reaction Rate Constants (Evan Komp and Stephanie Valleau)

1,517,419 个量子反应速率常数积 kQM(T)QR(T) 的数据集，根据模型单势垒和双势垒最小能量路径的传输系数计算得出。这里 kQM(T) 是温度 T 下的量子反应速率常数，QR(T) 是使用刚性转子和谐振子近似计算的反应物配分函数。该数据集是为 Ref [1] 创建的，用于训练和测试 DNN 以预测 logkQM(T)QR(T)。有关详细信息和数据集，请参见 zenodo 网页 https://zenodo.org/record/5510392#.YUkbjWZKhOc

A dataset consisting of 1,517,419 quantum reaction rate constant products kQM(T)QR(T), calculated using the transmission coefficients derived from the single-barrier and double-barrier minimum energy paths of the model. Here, kQM(T) denotes the quantum reaction rate constant at temperature T, while QR(T) is the reactant partition function computed via the rigid rotor and harmonic oscillator approximations. This dataset was constructed for Ref. [1] to train and test deep neural networks (DNNs) for predicting logkQM(T)QR(T). For further details and access to the dataset, please visit the Zenodo webpage at https://zenodo.org/record/5510392#.YUkbjWZKhOc

搜集汇总

背景与挑战

背景概述

该数据集包含约151.7万个量子反应速率常数积，基于单势垒和双势垒最小能量路径的传输系数计算得出，用于训练和测试深度神经网络以预测logkQM(T)QR(T)。它由华盛顿大学化学工程系于2021年发布，遵循CC BY 4.0许可，相关详细信息可在Zenodo网页获取。