Objective to improve traditional BP algorithm, to build effective the pattern recognition of the activity of medicament.
目的改进传统的BP算法，建立有效的药物活性识别模式。
For the shortages of the method in which network flux is predicted by forward neural network based on traditional BP algorithm, a 2-level network flux predict-adjust model is constructed.
针对基于传统BP算法的前向神经网络预测网络流量方法的不足，构建了一种二级的网络流量预测-校正模型。
To improve the networks'classification effect and train speed, the additive momentum and self-adaptive–study-rate adjustment method are adopted further to improve traditional BP algorithm.
为了提高网络的分类效果以及训练速度，采用了附加动量法和自适应学习速率调整法对BP算法进行了改进。
应用推荐