molecular subgraphs),multiple linear(regression) equation was established,of which the correlation coefficient is 0.
以硝基烷烃分子结构中不同基团作为描述码,以每一个描述码作为相应的分子子图项,进行了多元线性回归,预估多硝基烷烃化合物的生成焓,取得了较好的结果,其回归方程相关系数达到0。
On the basis of the molecular subgraph coding method,the molecular subgraph code of alkanes is treated as the input parameter of artificial neural network to predict heat of formation of alkanes.
采用分子子图编码方法将烷烃的分子子图码作为人工神经网络(ANN)的输入参数 ,对烷烃的生成焓进行预测 ,取得了满意的结果 ,其拟合方程的回归系数达到 0 。
On the basis of molecular subgraph coding method, a back propagation neural network was trained on the molecular subgraph code to predict Gibbs energy of alkanes.
采用分子子图编码法将烷烃的分子子图码作为人工神经网络的输入参数 ,对烷烃的吉布斯自由能进行预测 。