Closed jiayihu closed 3 years ago
The dataset is not very much a normal distribution. Original dataset on the left, generated samples for rainfall and flow on the right. Flood-related data is expected to contain spurious samples, which are not in the normal distribution by definition.
However, for the purpose of the thesis, it shouldn't be an issue generating data concentrated in the mean and standard deviation. In the worst case, I can pick the test dataset and pick random values from it
Epic #61