Search Results

The cloud monitoring tools is terrible, but there's some pointers here: https://developers.google.com/earth-engine/guides/monitoring_usage#view_consumption_in_cloud_monitoring The important part is th …

Reducers weight the pixels according to the image mask; addition doesn't. You can see in the inspection that the input pixels and the final pixel have a mask of approximately 81%. 724*0.81 = 586. If …

You didn't set a scale in the Export. You also haven't set a neighborhood on the fastDistanceTransform() call, which means you're only getting a default neighborhood of 256 pixels, which at 100m scale …

Theres a lot of land with slope = 0 at 10m. If you don't want water, you need to use some other dataset for determining water/not-water.

No part of the example you gave involves any significant caching on the server side. There is no unique ID to be had; nothing's kept in that example. The only things that are kept longer than a few se …

As the error messagbe says: you've provided an incorrect argument to Classifier.train(). The inputProperties argument is expected to be a list of property names. That's not what you used. You proba …

FilterBounds includes any image whose footprint intersects your ROI in any way. That doesn't mean there are always valid pixels in your ROI, and if there are no valid pixels, then the output of a red …

Add the bands from ee.Image.pixelLonLat() to your image. But getInfo is not the right way to get the pixels. Check out ee.data.computePixels(). It'll even give you numpy format if you ask for it.

You can figure out which of the operations is too big by trying to print the size of each of the point samples, and the classifier.explain(), and seeing which one barfs first. Most likely, you are sam …

You can skip the If() by just merging the collections and taking the first().

It's expensive. You need to make a histogram of all the pixels in each image (mapped reduceRegion using a frequencyHistogram reducer) and store the results (just the list of values found, not the coun …

No, there's nothing built in for that. It would make a good feature request though.

It's no big deal if you call it twice. Most of the library notices that it's already been initialized once and skips doing unnecessary work.

getAsset just gets metadata about the asset; it doesn't actually load it. In order to use it, you use the FeatureCollection constructor with the asset's ID/path. ee.FeatureCollection("projects/ee-alen …

When passed a Geometry, Feature or FeatureCollection, Map.addLayer uses collection.draw() to create the layer. There aren't a lot of options to draw, so instead, you should make your own call to col …