I tried to train a forest-based classification and regression model, to predict a binary categorical raster. When I trained the model, I choose to use 12 bands from Sentinel-2 satellite images as the Explanatory training raster, but it showed that these 12 bands had an error-110180, which described the problem is that at least one field in the Explanatory Training Variables parameter contains records, and at least 95 percent of those records have the same value. More variability in the values is required.
However, I ran this train model process in another project, with the same data with the same processing, and sometimes it succeed, which made me quite confuse where the problem is.
