Skip to main content

All Questions

Filter by
Sorted by
Tagged with
0 votes
0 answers
22 views

What is the optimal number of samples in a random forest classification?

I am conducting a supervised classification as part of a LULC analysis. I have 8 classes with sample sizes maxing at 500. However, I am still seeing some misclassification and want to add more classes,...
Ceiba's user avatar
  • 1
1 vote
1 answer
233 views

Is there a way to obtain the variable importance for a specific class in Earth Engine?

I used a .smileRandomForest() and I know that by calling .explain() you can see the relative importance of the variables for the general classification... but I would like to know which inputs ...
M_LeonPerez's user avatar
2 votes
0 answers
599 views

"User memory limit exceeded" during Random Forest classification in GEE

I succeeded in an initial attempt at using Random Forest classification on my input image (an exported mosaic covering ~ 2,797 km2). I used 30 trees, classified based on 4 bands, and passed a training ...
Jay Logan's user avatar
0 votes
1 answer
2k views

Normalising image bands in google earth engine before classification [closed]

I have an image that has four bands. Three of them are derived from Sentinel-2 to give NDVI, NDBI and SAVI. The fourth band is derived from a Sentinel 1 SAR image collection. The Sentinel 2 bands have ...
Anthony W's user avatar
  • 255
3 votes
0 answers
3k views

Encoding categorical variable for random forest with sklearn

I have a random forest model that works pretty well, taking a bunch of vanilla remote sensing raster data as input. I think it could be improved with addition of some information that I currently have ...
LAT's user avatar
  • 579
0 votes
1 answer
1k views

Classify polygons with Random Forest on Python

I have a segmentation shapefile made with e-cognition containing many polygons of which a part classified for the train file. I would like to classify them by applying labels (e.g. water, vegetation, ...
vins_26's user avatar
  • 443
1 vote
0 answers
389 views

Using output of unsupervised cluster analysis to train supervised classification for vegetation mapping?

I'm using Google Earth Engine to classify mangrove forests from Landsat imagery to calculate extent. Let's say I perform an unsupervised K-means cluster analysis on a Landsat scene to identify general ...
Samual Chance's user avatar