Charging pile sales volume model training method and device, equipment and storage medium

The invention discloses a charging pile sales volume model training method and device, equipment and a storage medium, and relates to the field of neural networks, and the method comprises the steps: obtaining a data set containing charging pile sales data, dividing the data set into a training set and a test set, and carrying out the initialization processing of the test set and the training set; constructing a charging pile sales volume model network structure including an LSTM network and a GRU network, and performing iterative training on the model based on the initialized test set and training set; wherein the LSTM network and the GRU network form a two-time stacking structure; and obtaining a candidate super-parameter group number conforming to the model through cross validation, and selecting a target super-parameter combination according to the performance index to obtain a charging pile sales volume model. According to the scheme, sales volume prediction is carried out based on an LSTM-GRU prediction