python - SciKit-Learn: Grid Search CV Recording Times? -
i'm doing basic machine learning , want grid search xgboost (to find best hyperparameters). here's i'm doing:
from sklearn.metrics import classification_report sklearn.pipeline import pipeline sklearn.grid_search import gridsearchcv import xgboost xgb pipeline = pipeline([('clf', xgb.xgbregressor(silent=false))]) parameters = { 'clf__n_estimators': (500, 1000), 'clf__learning_rate': (0.01, 0.05, 0.1), 'clf__max_depth': (10, 20, 30) } grid_search = gridsearchcv(pipeline, parameters, n_jobs=-1,verbose=2, scoring='neg_mean_absolute_error') grid_search.fit(x, y)
so start getting output in jupyter notebook follows:
fitting 3 folds each of 18 candidates, totalling 54 fits [cv] clf__max_depth=10, clf__learning_rate=0.01, clf__n_estimators=500
all seems goes through of combinations , doesn't terminate. instead, starts printing following:
[cv] clf__max_depth=10, clf__learning_rate=0.01, clf__n_estimators=500 -119.5min [cv] clf__max_depth=10, clf__learning_rate=0.01, clf__n_estimators=500 -119.8min
what mean? expecting grid search perform 54 fits , done seems doing other stuff afterward?
thanks help!
Comments
Post a Comment