I want to predict forest biomass over my study region. To do that I have trained Tensorflow model using field data (70 Temporary forest plots) and some remote sensing variables that i've extracted from images disponible in the Google Earth Engine Dataset. Thus, for each plot (training pixels) I got the following training data (sample).

sampling plan

Training Data

After that I have trained a tensorflow model (DNN) that predict the output variable (acc_carbomasse) given the other input variables.

My question is : How can I apply my trained model on ee.ImageCollection in order to predict acc_carbomasse variable over my full study region with the possibility of mapping that variable ?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.