Installing xgboost package

Autonomous Machines Jetson & Embedded Systems Jetson Nano
#1

Hi,
I was trying to install the package xgboost on my nano board, and its failed.
I tried command python3 -m pip install xgboost but the command fails with error message:
Failed building wheel for xgboost
I have Python version 3.6.9
and pip3 version 20.1

I have also tried following the instructions from https://gist.github.com/DanielBeckstein/932087ee116cc2f72bfc6e3e078e899d and failed since the cmake version was less than 3.13.
So if anyone has any idea to try it will be helpful.
Thanks
Ashik P.

#7

Hi ashik.p,

We haven’t try xgboost, not sure if can work, may other developers share experiences if they done similar.

#8

I am trying to train an XGBoost model on my WSL2-Ubuntu 20.04 setup. However, I get an error on my local machine when I set ‘tree_method’: ‘gpu_hist’ in params:

"Exception in gpu hist: NCCL failure : unhandled system error"

I can fit a model using ‘tree_method’: ‘hist’ although is now fitting on the CPU rather than the GPU, which defeats the purpose of using RAPIDS+Cuda!

Note as part of your DLI course FUNDAMENTALS OF ACCELERATED DATA SCIENCE WITH RAPIDS, I was able to train an XGBoost model on the GPU cluster, i.e. with ‘tree_method’: ‘gpu_hist’ in params.

Any suggestions?