First steps with Geemap.
Date:
A ready to use short demonstration of several possibilities with Geemap. The objective is to show that, with minimum effort, it’s possible to create a model that estimates the population count for a given lat/lon and create an interactive map to visualize the results.
–>
Train data is on the left and test data on the right. A Decision Tree Regressor is trained at the pixel level, to estimate the population count, given by the Worlpop dataset. On the test side, we observe some a radically different spatial pattern: the model produce a higher resolution map that seem globally consistent with the background high resolution optical imagery. However there are also a high proportion of error predictions (with high population count in marsh for examples …)