I want to make a binary clasificación (1 = burned area, 0 = non burned area) using smile random forest, to train Sentinel 2 SR (12 band included NDVI and DBR) and get new burned trained areas.

I try to make IR with feature collection:

var buffer = function(feat){return feat.buffer(1000,50)};

var nonarea = area.geometry()

Map.addLayer(nonarea,{palette: ['blue']},'Área de no concurrencia de incendio');

var nonocurrence = post_cm_mos_AD.sample({
                        region: nonarea,
                        scale: 30,
                        tileScale: 16
                          return feature.set('clase',ee.Number(0))

but the console return Computed value is too large

the, I haven been trying to make it, using stratifiedSample, sample code:

var first_img = b_binary;
//Map.addLayer(first_img, vis,'Áreas de incendio para entrenar');

var training_burn2 = post_cm_mos_AD.stratifiedSample({
                                numPoints: 10,
                                classBand: "constant",
                                region: area,
                                tileScale: 16,
                                geometries: true,

But the message in the console is: Computation timed out

This is my link code; https://code.earthengine.google.com/90f4f15023f471059a06ffd57585ea4f

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.