XGBoost Classification Algorithm in Machine Learning | Python | XGBoost Algorithm

[@shwetaredkar734]

The best tutorial so far. This video definitely needs more views and likes. What if we had to do RandomsearchCV or GridSearchCV with 10fold CV to hyper tune the parameters and fit the model than how should we do it?

[@StatsWire]

Thank you so much for the kind words. To do RandomsearchCV or GridSearchCV we can define them and pass it to the function. Here is one example # RandomizedSearchCV params = { 'min_child_weight': [1, 5, 10], 'gamma': [0.5, 1, 1.5, 2, 5], 'subsample': [0.6, 0.8, 1.0], 'colsample_bytree': [0.6, 0.8, 1.0], 'max_depth': [3, 4, 5] } folds = 3 param_comb = 5 skf = StratifiedKFold(n_splits=folds, shuffle = True, random_state = 1001) random_search = RandomizedSearchCV(xgb, param_distributions=params, n_iter=param_comb, scoring='roc_auc', n_jobs=4, cv=skf.split(X,Y), verbose=3, random_state=1001 ) random_search.fit(X, Y) y_test = random_search.predict_proba(test) similarly you can do for GridSearchCV

[@lavanyareddy406]

could you share us the link for the dataset.

[@StatsWire]

Could you please tell me the dataset name? I can't see my own videos.

[@abhishekarya2628]

@@StatsWire census-data.csv

[@hakanunal4565]

You are great my friend :)

[@StatsWire]

Thank you:)

[@nasifosmanshuvra8607]

Thanks for the tutorial.. If I have a large Image dataset and I use data generator to generate batches of images. How can I pass then all the images to xgb classifier to fit the model?

[@StatsWire]

You need to convert those images to numbers using numpy array

[@mohammedzia1015]

We should not use XGBoost for Image data. XGBoost works better for Tabular data. For image data, we have Deep Learning techniques.

[@StatsWire]

@@mohammedzia1015 Yes for image data you can use CNN

[@mazharalamsiddiqui6904]

Nice tutorial

[@StatsWire]

Thank you

[@mohammedzia1015]

Is not education an important feature to predict Income? Why can't you do One Hot Encoding for Education column and use it?

[@StatsWire]

Education is an important feature in many datasets and this video is just for demo purposes on how to apply xgboost. You can play around with features and check the accuracy of the model and select the final one.