The ML neural network will output values / probabilities for each of the tropical cyclone categories at each grid point. We will want to translate this to the centre points of the storms. This can be done by filtering to find all the grid points above a certain probability and then finding the centroid. This is complicated by the fact that there may be multiple TCs at a given time, so there will need to be some way to separate out multiple storms
The ML neural network will output values / probabilities for each of the tropical cyclone categories at each grid point. We will want to translate this to the centre points of the storms. This can be done by filtering to find all the grid points above a certain probability and then finding the centroid. This is complicated by the fact that there may be multiple TCs at a given time, so there will need to be some way to separate out multiple storms