CODEWITHSUNDEEP
pythonxgboost
Slavka
Slavka

Reputation: 1148

mape eval metric in xgboost

i'm trying to use MAPE as eval metric in xgboost, but get strange results:

def xgb_mape(preds, dtrain):
   labels = dtrain.get_label()
   return('mape', np.mean(np.abs((labels - preds) / (labels+1))))

xgp = {"colsample_bytree": 0.9, 
   "min_child_weight": 24, 
   "subsample": 0.9, 
   "eta": 0.05, 
   "objective": "reg:linear", 
   "seed": 70}

cv = xgb.cv(params = xgp, 
        dtrain = xgb.DMatrix(train_set[cols_to_use], label=train_set.y),
        folds = KFold(n = len(train_set), n_folds=4, random_state = 707, shuffle=True),
        feval = xgb_mape,
        early_stopping_rounds=10,
        num_boost_round=1000,
        verbose_eval=10,
        maximize=False
        )

It returns:

[0]     train-mape:0.780683+0.00241932  test-mape:0.779896+0.0024619
[10]    train-mape:0.84939+0.0196102    test-mape:0.858054+0.0184669
[20]    train-mape:1.0778+0.0313676     test-mape:1.10751+0.0293785
[30]    train-mape:1.26066+0.0343771    test-mape:1.30707+0.0323237
[40]    train-mape:1.37713+0.0347438    test-mape:1.43339+0.030565
[50]    train-mape:1.45653+0.042433     test-mape:1.52176+0.0383677
[60]    train-mape:1.52268+0.0386395    test-mape:1.5909+0.0353497
[70]    train-mape:1.5636+0.0383622     test-mape:1.63482+0.0301809
[80]    train-mape:1.59408+0.0378158    test-mape:1.66748+0.0315529
[90]    train-mape:1.61712+0.0403532    test-mape:1.69134+0.0325177
[100]   train-mape:1.63028+0.0389446    test-mape:1.70578+0.0316045
[110]   train-mape:1.63556+0.0375842    test-mape:1.71153+0.031564
[120]   train-mape:1.63509+0.0393198    test-mape:1.7117+0.0320471

Train and test results increases with maximize=False, also early_stopping doesnt work properly. Where is error?

UPD. added -1* to xgb_mape, it solved problem. Looks like maximize parameter doesn't work properly for custom eval functions.

Upvotes: 7

Views: 9582

Answers (1)

data_steve
data_steve

Reputation: 1648

According to this xgboost example of implementing Average Precision metric, since the xgb optimizer only minimizes, if you implement a metric that maximizes, you have to add a negative sign (-) in front of it, like so:

def pr_auc_metric(y_predicted, y_true):
    return 'pr_auc', -skmetrics.average_precision_score(y_true.get_label(), y_predicted)

So yours would be:

def xgb_mape(preds, dtrain):
   labels = dtrain.get_label()
   return('mape', -np.mean(np.abs((labels - preds) / (labels + 1))))

Upvotes: 3

Related Questions