CODEWITHSUNDEEP
rxgboost
Kumar Divyesh
Kumar Divyesh

Reputation: 89

Xgboost error : label must be in [0, num_class), num_class=2

Getting this error while implementing XGboost for the titanic problem

Error in xgb.iter.update(bst$handle, dtrain, iteration - 1, obj) :
[03:26:03] amalgamation/../src/objective/multiclass_obj.cc:75: Check failed: label_error >= 0 && label_error < nclass SoftmaxMultiClassObj: label must be in [0, num_class), num_class=2 but found 2 in label.

Following is my code:

#Parameter  ie no of class
nc <- length(unique(train_label))
nc
xgb_params <- list("objective"="multi:softprob",
                       "eval_metric"="mlogloss",
                       "num_class"=nc)
watchlist <- list(train=train_matix,test=test_matix)


#XGB Model
bst_model <- xgb.train(params = xgb_params,data = train_matix, nrounds = 100,watchlist = watchlist)

How can I resolve this?

Upvotes: 3

Views: 21137

Answers (3)

MartiTro
MartiTro

Reputation: 31

try xgb.XGBRegressor(use_label_encoder=False) instead of XGBClassifier().

Upvotes: 0

Nish
Nish

Reputation: 61

I solved it like this. My class labels were -1, 0 and 1. So my num_class=3. I had to increment class labels by 1 in order to compatible with the range [0,3). Note that in this range 3 is excluded and valid labels are 0, 1, 2. So my converted class labels were 0,1,2.

In addition I change the code to use for multiple class classification.

objective has been changed to 'multi:softmax' and 'num_class' param is added.

xgb1 = XGBClassifier(
    learning_rate=0.1,
    n_estimators=1000,
    max_depth=5,
    min_child_weight=1,
    gamma=0,
    subsample=0.8,
    colsample_bytree=0.8,
    objective='multi:softmax',
    nthread=4,
    scale_pos_weight=1,
    seed=27,
    num_class=3,
    )

In modelfit() finction 'auc' was replaced with 'merror'

def modelfit(alg, dtrain, predictors, useTrainCV=True, cv_folds=5, early_stopping_rounds=50):
if useTrainCV:
    xgb_param = alg.get_xgb_params()
    #change the class labels
    dtrain[target] = dtrain[target] + 1
    xgtrain = xgb.DMatrix(dtrain[predictors].values, label=dtrain[target].values)

    cvresult = xgb.cv(xgb_param, xgtrain, num_boost_round=xgb_param['n_estimators'], nfold=cv_folds,
                      metrics='merror', early_stopping_rounds=early_stopping_rounds)
    alg.set_params(n_estimators=cvresult.shape[0])
    print(cvresult.shape[0])

# Fit the algorithm on the data
alg.fit(dtrain[predictors], dtrain[target], eval_metric='merror')

# Predict training set:
dtrain_predictions = alg.predict(dtrain[predictors])

# Print model report:
print("\nModel Report")
print("Accuracy : %.4g" % metrics.accuracy_score(dtrain[target].values, dtrain_predictions))

Upvotes: 6

Eran Moshe
Eran Moshe

Reputation: 3208

I'm guessing you have a multi label problem. Since there's no example I can only guess you should pass nc+1 instead of nc. Good luck.

Upvotes: 2

Related Questions