[Dclib-devel] Random forest model: DLIB_ASSERT(trees[i].size() > 0, "A tree can't have 0 leaves.");

[Vishal K. S. <vis...@gm...>]

Hi,

I am trying to train random forest regression model.

I am getting an error.

*DLIB_ASSERT(trees[i].size() > 0, "A tree can't have 0 leaves.");*

Please find the code below:

        std::vector<sample_type> samples;
std::vector<double> labels;

*// Here I fill the dataset;*

cout << "Starting... " << endl;
random_forest_regression_trainer<dense_feature_extractor> trainer;
trainer.set_num_trees(100);
trainer.set_seed("random forest");
cout << "Training..." << endl;
auto df = trainer.train(samples, labels); * // While training it crashes.*
cout << "Saving the model..." << endl;
serialize("model.dat") << df;
random_forest_regression_trainer<dense_feature_extractor>::trained_function_type
df1;
cout << "Loading the model..." << endl;
deserialize("model.dat") >> df1;

Can anyone point out to what I am doing wrong ?

 Any help will be highly appreciated.

Best Regards,
Vishal.