In [1]:

    

import numpy as np
import scipy.sparse
import pickle
import xgboost as xgb
import time


    

In [7]:

    

ls ../data/mushroom/


    

    




ACE2.jpg  Agaricus_californicus(P).jpg            agaricus.txt.test
ACE.jpg   Agaricus campestris 1 Michael Beug.jpg  agaricus.txt.train
ACP.jpg   Agaricus_campestris(E).jpg              Agaricus_xanthoderma(P).jpg

In [2]:

    

dtrain = xgb.DMatrix('../data/mushroom/agaricus.txt.train')
dtest = xgb.DMatrix('../data/mushroom/agaricus.txt.test')


    

In [3]:

    

param = {'max_depth':2, 'eta':1, 'silent':1, 'objective':'binary:logistic' }


    

In [6]:

    

watchlist  = [(dtest,'eval'), (dtrain,'train')]
num_round = 2
start = time.clock()
bst =xgb.train(param, dtrain, num_round, watchlist)
print time.clock() - start


    

    




0.019947






    




[0]	eval-error:0.042831	train-error:0.046522
[1]	eval-error:0.021726	train-error:0.022263

In [7]:

    

preds = bst.predict(dtest)


    

In [8]:

    

preds


    

    
Out[8]:





array([ 0.28583017,  0.92392391,  0.28583017, ...,  0.92392391,
        0.05169873,  0.92392391], dtype=float32)

In [ ]: