基于后向气团轨迹的大气污染特征时序混合模型研究

许睿, 刘相阳, 文益民, 沈世铭, 李建

Research on time series hybrid model of air pollution characteristics based on backward air mass trajectory

Rui Xu, Xiangyang Liu, Yimin Wen, Shiming Shen, Jian Li

表2 神经元数量对H?VGG?LSTM模型性能的影响

Table 2 Effect of the number of neuron nodes on the performence of H?VGG?LSTM

Neurons RMSE MAE 16 0.941 0.756 32 0.584 0.523 64 0.763 0.654 128 0.836 0.736