Open JonoDW opened 6 years ago
Could you post the files you used for training and prediction somewhere to reproduce?
The files are attached. Taken from the "iris" dataset.
The commands used were as follows ./ranger --verbose --file iris_data/iris_train --depvarname Species --treetype 1 --write ./ranger --verbose --file iris_data/iris_test --predict ranger_out.forest
Thanks
Thanks. Currently only numeric values are supported in the C++ version. Please code the Species as numeric. We should change the C++ version to accept non-numerics (I've changed the title).
Thanks for the feedback. It works with your advice (encoding the depvar into numeric format)
Hi
I've noticed that the C++ standalone version outputs a file of "0"s where the classifications should be. I've tried the C++ example given in the paper "A Fast Implementation of Random Forests for High Dimensional Data in C++ and R" with the same result. Is this a bug or am I doing something wrong?