Skip to content
Branch: master
Find file Copy path
Find file Copy path
2 contributors

Users who have contributed to this file

@tqchen @LevineHuang
executable file 26 lines (18 sloc) 886 Bytes
import numpy as np
import scipy.sparse
import xgboost as xgb
### simple example for using external memory version
# this is the only difference, add a # followed by a cache prefix name
# several cache file with the prefix will be generated
# currently only support convert from libsvm file
dtrain = xgb.DMatrix('../data/agaricus.txt.train#dtrain.cache')
dtest = xgb.DMatrix('../data/agaricus.txt.test#dtest.cache')
# specify validations set to watch performance
param = {'max_depth':2, 'eta':1, 'silent':1, 'objective':'binary:logistic'}
# performance notice: set nthread to be the number of your real cpu
# some cpu offer two threads per core, for example, a 4 core cpu with 8 threads, in such case set nthread=4
watchlist = [(dtest, 'eval'), (dtrain, 'train')]
num_round = 2
bst = xgb.train(param, dtrain, num_round, watchlist)
You can’t perform that action at this time.