Questions tagged [random-forest]
An ensemble classifier that consists of many decision trees and outputs the class that is the mode of the classes output by individual trees
151
questions
0
votes
0
answers
22
views
Use two different composite images in Random Forest land cover classifications [closed]
My goal is to create land cover maps through a supervised random forest classification using GEE. The main land cover class I am classifying is mangroves, as well as more secondary land cover classes ...
- 1
0
votes
0
answers
41
views
Estimating forest cover density through Google Earth Engine
I am trying to make a forest cover map (estimating forest canopy density) of a particular district and block. But I don't know where to start since I am a beginner. I have researched and came to a ...
0
votes
0
answers
33
views
Exporting classified land cover class as an asset for NDVI calculations
I have conducted a supervised classification in Google Earth Engine using Random Forest, to which I have provided the code.
https://code.earthengine.google.com/3b3d4583f26a4651f4fbe10aa80c5f30
Map....
- 1
0
votes
1
answer
41
views
Exporting raster to my assets
I am working on a vegetation type classification using random forest and Sentinel-2 imagery, and I would like to export my final output map (GeoTIFF) which contains 7 classes and some masked areas ...
- 1
0
votes
0
answers
17
views
What do the values in trees of Random forest GEE mean?
I get these trees in GEE:
trees: List (29 elements) 0: n= 116838
node), split, n, loss, yval, (yprob)
* denotes terminal node
1) root 116838 31180 8 (0.010997 0.0066155 0.019316 0.013984 0.012195 ...
- 15
0
votes
2
answers
38
views
Error while trying to train RandomForest in R
I have this data:
diff is the results of image difference from bitemporal imagery, and 2, 4, 6, and 8 is the bands of PlanetScope imagery.
When I trying to train this data using RandomForest, ...
- 17
0
votes
1
answer
29
views
Random forest classification on QGIS gets stocked at 66% with an error message whose conditions have been met
I am running a random forest classification on QGIS using dzesaka pluggin, after picking my training sites and running the classification,
it gets to 66% and a prompt message appears that i ensure ...
0
votes
0
answers
21
views
variablesPerSplit default value rounding in smileRandomForest()
I have used the Google Earth Engine's smileRandomForest() classifier and leaving the variablesPerSplit to the default value sqrt(N_variables) with N_variables=8 input variables, which I think is ...
0
votes
0
answers
30
views
Get working of Random Forest in GEE in understandable form
I used this code to see decion tree in RF GEE:
var decisionTrees = ee.List(trained.explain().get('trees'))
print("Tree decision", decisionTrees)
I get this:
Tree decision
List (29 elements)
...
- 15
0
votes
0
answers
109
views
GEE - random forest doesn't classify my whole image
I'm new to GEE and I'm trying to do a machine learning supervised classification using random forest. it turns out that I made a code taking different parts of other codes and it works well for small ...
1
vote
0
answers
60
views
Export table to drive "User memory limit exceeded" on GEE
I'm exporting the table from the random forest model result, showing the relationship between the predicted values and the observed values in terms of the training data. However, when I ran the task ...
- 31
0
votes
1
answer
23
views
Class not recognized in dataset in RF classification with GEE
I want to classify imagery from different regions with a classifier trained in one specific region.
My training and test dataset contains polygons which are assigned to 4 different classes (class 0 - ...
- 17
0
votes
0
answers
52
views
Random forest regression error on GEE "No valid training data were found."
I'm using RF regression model on GEE. The output can be successfully generated but when I tried to make a "Variable Importance Histogram", the error happened and said "Dictionary (Error)...
- 31
0
votes
1
answer
48
views
Random forest error on GEE
I have a problem with random forest classification, receiving an error message after inputting the training dataset. My code includes a long preprocessing for Sentinel1 and Sentinel2, but here I'll ...
- 31
1
vote
2
answers
83
views
Random forest/Maxent environmental predictor layers using vector data, in GEE
I am attempting to use a vector layer as a predictor environmental layer for use in classification/predictor models (eg Random Forest/Maxent). I have attempted to include a bathymetric substrate layer,...
- 21
0
votes
0
answers
24
views
Set random as a specific value for classification in GEE
How do I set random as a specific value for classification in GEE?
Random forest classification in GEE works when I use >= or <= in these expressions:
var trainingTesting = classifierTraining....
- 15
0
votes
0
answers
74
views
GEE error, User memory limit exceeded
I am doing a supervised classification and I get the error that User memory limit exceeded, however my data is not that complex anymore. I exported my inputPropeties as a stacked layer, I am doing the ...
- 93
1
vote
1
answer
121
views
What formula does GEE use for calculation of Importance in Random Forest?
I got such values of importance in GEE, using such code
var variable_importance = ee.Feature(null, ee.Dictionary(trained.explain()).get('importance'));
What formula does GEE use for calculation it? ...
- 15
1
vote
0
answers
68
views
How to apply GEE training data to a new region
The link to my code is included here but I will paste it as well: https://code.earthengine.google.com/c9b3aec923007e896a3a079446001dac
I am trying to dynamically classify ground cover using supervised ...
0
votes
1
answer
36
views
Wrong Palette of color when doing a supervised classification GEE
I am doing a classification of forest and nonforest area using the random forest classifier. At the end of the process I am not getting the color separation, the classified image looks like this:
It ...
1
vote
2
answers
129
views
Loop rename rasters in brick or stack in R
I'm running a randomforest on a raster stack/brick in R. I know how to rename individual layers in the stack, but since my stacks have numerous Landsat scenes, I was hoping to loop through the stack. ...
- 385
0
votes
0
answers
22
views
Raster values NoData category not recognized as factor by caret random forest model
I want to predict a Random Forest model onto the values from a raster stack of 20 different variables, some of which are categorical and treated as factors in the model. I uses values() to extract ...
- 63
0
votes
0
answers
45
views
Randomforest Land Cover Classification Band Error? Image.blend Error on Earth Engine
I've been trying to find the error in this specific code for the past few weeks but I've genuinely been stumped. Here's a link to a sharable snapshot of the current code: https://code.earthengine....
- 1
0
votes
0
answers
51
views
Predict caret random forest model onto raster stack fails due to handling of categorical variables
Using ArcGIS Pro and R I made a random forest model to predict the concentration of a groundwater contaminant in wells. Some of my explanatory data is categorical (like soil drainage and land use) so ...
- 63
0
votes
0
answers
20
views
Layer error: Description length exceeds maximum [duplicate]
for my master's thesis work, I am doing a crop-type mapping using supervised classification. The code is working. However, I cannot view the classification map as I got this error 'Layer error: ...
- 93
0
votes
1
answer
44
views
Low accuracy for addition of more predictor bands in Random forest classifier
Hi there Im working on a project for estimation Above ground Biomass in hilly terrain for mature forests and asscessing accuracy of different classifiers availiable in GEE such as Random forest and ...
0
votes
1
answer
144
views
Error in GEE: Property of feature is missing
I am trying to do a supervised classification in GEE, I tried the same code over one area of interest and it worked. However, when I combined both sites, an error occurred "Property '0_SR_B2' of ...
- 93
0
votes
1
answer
100
views
GEE_No valid training data
I am trying to conduct supervised classification using RF, however I face problems with my training data. I have my data as a point shp file, however, whenever I use it, it gives me the error "no ...
- 93
0
votes
1
answer
46
views
Different no. of training points than the one I am asking for
Does anyone know, what could be the reason, I am getting different numbers of my training samples than the one I am asking in the code? Like in my example, for crop no.6, I am asking for 200 points, ...
- 93
1
vote
0
answers
55
views
Use shapefile with multiclass as a training data
How can I use a shapefile containing different land cover classes as my training data in GEE?
Here is the code I am using,however there is always an error with the classification:
https://code....
- 93
0
votes
0
answers
112
views
Increasing accuracy with RF classifier in GEE
I am doing a supervised classification using RF classifier in GEE. However, the accuracy that I am getting is not exceeding 0.67, even if I increase the training sample or no. of trees in my ...
- 93
1
vote
1
answer
50
views
Missing a couple of tiles after classification, however it was shown in the stacked Landsat image
I am trying to do crop-type mapping using RF in GEE.the script is working, however, part of the final classified layer (a couple of tiles) is missing in the end product which I do not know why! I even ...
- 93
0
votes
1
answer
804
views
How to overcome "Output of image computation is too large" error
Does anyone know how to avoid the error of "Output of image computation is too large" in supervised classification using RF? I even tried up to 16 tilescale but it still does not work. Noted ...
- 93
0
votes
1
answer
233
views
Error: Property 'class' of feature '0_0' is missing
Using GEE, I am trying to do a crop-type classification using Random Forest. My training data are two shapefiles I uploaded to GEE and made a FeatureCollection from their geometries. However, it gives ...
- 93
1
vote
1
answer
103
views
Merging training data from two different shapefiles
Using GEE, I am trying to do a crop-type classification using Random Forest. My training data are two shapefiles I uploaded to GEE and made a FeatureCollection from their geometries. Here is the ...
- 93
0
votes
1
answer
46
views
Stacked image created, while most of the bands became masked
I am trying to do a crop-type classification, so I created a stacked image from my collection for training by RF. However, when the collection is stacked, most of the bands inside the stacked image ...
- 93
1
vote
1
answer
165
views
Using RF as a classifier, however no valid training data founded?
I am new to GEE. I'm trying to do a crop mapping for my ROI using Random Forest.
I am using Landsat8, computed index (NDVI, NDWI). Stacking all images of interest of Landsat with the needed bands, ...
- 93
0
votes
0
answers
28
views
Faster method to not consider NaN during RF classification?
To classify the pixels of a Sentinel-2 image I am using a Random Forest model.
Firstly, I am reading and stacking bands and indices pré-calculated from Sentinel-2 data (19 rasters in total with size ...
1
vote
1
answer
193
views
How to export decision trees from GradientTreeBoost in Google Earth Engine?
I'm running the Google Earth Engine examples for ee.Classifier.smileGradientTreeBoost and ee.Classifier.smileRandomForest, which are basically the same but with different training methods.
I was ...
- 11
0
votes
0
answers
62
views
Quantile regression forests in Google Earth Engine
I would like to use quantile regression forests in Google Earth Engine. Only random forests is available. Is there a way to run quantile regression forests as in R (e.g. ranger), but in GEE to take ...
- 1
1
vote
1
answer
629
views
Export too large issue in Google Earth Engine
Here is the path to my script- https://code.earthengine.google.co.in/?accept_repo=users%2Fujavalgandhi%2FEnd-to-End-GEE&scriptPath=users%2FchakrabortySoumik95%2Ffirstprogram%3Asentinel%20Australia%...
1
vote
0
answers
40
views
Is it necessary to have the same spatial resolution of all the bands of an image before classifying the image? [duplicate]
Is it necessary to have the same spatial resolution of all the bands of an image before classifying the image? In my case, I am working with the 10m and 20m spatial resolution bands of Sentinel-2. I ...
0
votes
0
answers
50
views
Random forest and svm classification taking hours to run
I am performing crop classification using random forest technique on Sentinel-2 data. The file size of image on which classification has to be performed is only 26 mb.
I am using Jupyer notebook. Any ...
- 51
1
vote
1
answer
86
views
Fetch Probability from Random Forest Classified Image
I have performed random forest classification in Earth Engine would like to select only pixels with high probability. But I cant figure how it to be done. I see the probability value in the Inspector ...
- 11
1
vote
1
answer
110
views
Random forest classifcation for Sentinel-2 low out of bag but high prediction error
I used Random Forest for land cover classification on the Sentinel-2 images in a forest. I have 3 classes: forest (0), bareland (1), and cloud (2). My variables include 4 bands (B2, B3, B4, and B8), ...
- 349
0
votes
1
answer
37
views
Should I re-scale all the dataset to one scale before running them into model?
I have been working on the Flood susceptibility modelling and also have identified, collected and generated data of around 26 controlling factors(indicators). Furthermore, i'm going to put them into ...
1
vote
1
answer
134
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 ...
- 13
0
votes
1
answer
190
views
No training data found (random forest algorithm in GEE)
I was trying to map rice in the Indramayu Regency in Indonesia using the random forest algorithm. It has responded with no training data found.
When I used the same code previously for a different ...
- 11
0
votes
1
answer
632
views
Land cover classification using Google Earth Engine, and involving SAVI and EVI indices
I'm trying to do a land cover classification using Landsat-8 OLI/TIRS. However, I want to involve indexes (SAVI, and EVI) in this classification process. I don't know how to solve this problem. Is ...
0
votes
1
answer
248
views
How to add the bands value from CSV file to Image Collection for Classification
I am currently working with GEDI data and Random Forest Classification in Google Earth Engine. I have a csv file which contains 42 bands and about 3600 points. My aim is to import this csv into GEE ...