rcurtin
commented
2 years ago It seems like the silent=true option might help here (see #71); that worked at least partially for me, although I still have not managed to filter out things like:
[13:41:25] INFO: /workspace/srcdir/xgboost/src/tree/updater_prune.cc:101: tree pruning end, 12 extra nodes, 0 pruned nodes, max_depth=3
mlesnoff
Hello, I am using function xgboost of XGboost.jl. It seems that argument 'verbosity = 0' does not make 100% silent returns (there are printing messages anyway).
n = 50 ; p = 5 ; X = rand(n, p) ; y = rand(n) ; num_round = 3 ; fm = xgboost(X, num_round; label = y, max_depth = 2, verbosity = 0)Printing messages:
[1] train-rmse:0.280325 [2] train-rmse:0.266210 [3] train-rmse:0.255710 Booster(Ptr{Nothing} @0x000000001ecf3ea0)Is there a way to get a 100% silent return?