The recurrent neural network(RNN) model based on projective operator is studied.
研究了一种基于投影算子的神经网络模型。
A small discrepancy is due to the fact that the RNN only approximates the first-principles CSTR model.
一个小的差异是由于这样的事实，RNN CSTR采用的仅仅是近似模型。
As a biological neural mathematical model, RNN has particular advantages of associative memory, image processing and combinatorial optimization.
作为仿生神经元数学模型，随机神经网络在联想记忆、图像处理、组合优化问题上都显示出较强的优势。
First train RNN with datasets, then use trained RNN to provide the measurement of the outlyingness of data records. The performance of the RNNs is assessed by using a ranked score measurement.
先将数据集用于神经网络的训练，然后使用训练后的RNN对网络数据进行孤立度测量，根据度量结果判定是否为入侵行为。
Secondly, based on the recognition result, connected characters are segmented, and then recognized by a character RNN. Characters' posterior probabilities are calculated by a table-looking method.
然后，依据识别结果进行字符分割，使用字符RNN对分割后的字符进行识别，并利用查表法计算字符的后验概率；
Secondly, based on the recognition result, connected characters are segmented, and then recognized by a character RNN. Characters' posterior probabilities are calculated by a table-looking method.
然后，依据识别结果进行字符分割，使用字符RNN对分割后的字符进行识别，并利用查表法计算字符的后验概率；
应用推荐