jerrywn121
commented
1 year ago Thank you for your question. The dataset provided by the competition contains monthly data generated by different models (you can go to the contest website yourself to check it yourself). We use 36 because the input span is 12 months while the target to be predicted is 24 months. The start month of the input in the training set can be any month in a year. pred_length for SST is 26 because Nino is further obtained by convolution on the predicted SST with kernel size 3 in order to get 24-month Nino. Note that the convolution is not causal as it should have been, but empirically the performance is good
Hello, I'm a graduate student of geography science and I do some research on the prediction of ENSO recently. I want to ask how to set the sliding window of this training sample set? After looking at the code for a long time, I didn't understand this question. Does this competition use the data from January to December to predict the Nino3.4 index from January to December in the next two years? Why is the netCDF data of the Tianchi data set to month=36? If the sliding window does not start in January, how will the data conversion be handled? I noticed that this code set input_length=12 and pred_length=26, but I don't quite understand the purpose of this.