XGBoost Made Easy | Extreme Gradient Boosting | AWS SageMaker

Recently, XGBoost is the go to algorithm for most developers and has won several Kaggle competitions.
Since the technique is an ensemble algorithm, it is very robust and could work well with several data types and complex distributions.
Xgboost has a many tunable hyperparameters that could improve model fitting.
XGBoost is an example of ensemble learning and works for both regression and classification tasks.
Ensemble techniques such as bagging and boosting can offer an extremely powerful algorithm by combining a group of relatively weak/average ones.
For example, you can combine several decision trees to create a powerful random forest algorithm.
By Combining votes from a pool of experts, each will bring their own experience and background to solve the problem resulting in a better
outcome.
Boosting can reduce variance and overfitting and increase the model robustness.
I hope you will enjoy this video and find it useful and informative!
Thanks.
#xgboost #aws #sagemaker

[@mohamedsaber9634]

One of the best contents on the XGBoot subject. SIMPLE yet DEEP into details.

[@carsten7551]

I really enjoyed your video on XGBoost, Professor Ryan! This video made me feel much more comfortable with the model conceptually.

[@behradbinaei7428]

After searching 2 days , Finally I learned GB algorithms. Thank you so much

[@ahmadnurokhim4168]

This is exactly what I need, I see the other videos didn't cover the general concept like this

[@sirginirgin4808]

Excellent Explanation and to the point. Kindly keep up the good work Ryan.

[@WilsonJoey]

Great explanation of xgboost regression. Nice job professor.

[@mathsalmath]

Thank you Prof. Ahmed for a visual explanation. Great video.

[@robindong3802]

Thanks to Stemplicity, you make this profound algorithm easy to understand.

[@JIAmitdemwesen]

Very nice. I was quite confused in the beginning but the practical example help a lot to understand what is happening in this method.

[@maheshmichael6955]

Beautifully Explained :)

[@user-wr4yl7tx3w]

Great presentation. Clear and well explained.

[@johnpark7662]

Agreed, excellent presentation!

[@professor-ryanahmed]

Glad you liked it!

[@scottlapierre1773]

One of the best, for sure! Thank you.

[@ACTION206]

Very nice explanation

[@Ram-oj4gn]

wow great explanation..

[@SimbarasheWilliamMutyambizi]

Wonderful explanation

[@Yosoygarnold]

good explanation! thank you very much!.

[@professor-ryanahmed]

Glad it was helpful!

[@marcoaerlic2576]

Thanks for the great content, very well explained.

[@elchino356]

Great video!

[@professor-ryanahmed]

Glad you enjoyed it

[@khawarshehzad487]

Excellent video! loved the explanation

[@sudippandit1]

Your effort is great I really appreciate your efforts to make the things easy at a root level in this video. I would like to request to prepare one video like the same root level to make the idea of XGboost as easy as possible. How the Dmatrix, gamma and lambda parameters works to achieve the best model performance?

[@sarolovito2838]

Really excellent explanation!

[@mdgazuruddin214]

I think it's a tutorial on Gradient Boosting, Please make sure, and will be happy if you prove me wrong.

[@shrutichaubey2434]

great content

[@davidzhang4825]

Great video! Curios to know the difference between XGboost and Light GBM

[@HemanthGanesh]

Thanks much!!! Excellent explanation

[@fahadxraja2483]

sdf

[@aiinabox1260]

What youre saying is appllcable to Gradient boosting this is not xgboost .... You need to change the title as Gradient boosting .. xgboost u need to compute similarity score , gain & so on.

[@Mightyminiminds]

one of the best

[@aiinabox1260]

thanx for the fantastic explanation.... pl correct me if am wrong. my understanding is INITIAL model (average ) (A) -> residual -> Build an additional Tree to predict errors (B) -> with the combination of (A) & (B) it produces the target predicted value (P1); iteration 2 , this P1 (C) residuals -> predict errors (D) -> combination of C + D we get new predicted values...... Here the Tree B is called as weak learners and also called as Weak Learner. Am I correct ?

[@NadavBenedek]

The title says 'Gradient' but inside the video, where is the gradient mentioned?

[@jkho2085]

Hi, it is a wonderful contents on XGboost. I am a final year student and i wish to write it inside the report. However, it is hard to find the paper to support it.... Any suggestion?

[@theforrester2780]

Thank you, I needed this

[@thallamsairamya6843]

A novel xg boost tuned machine learning model for software bug prediction
We need a video regarding this exactly what I request
Plz make a video like that asap

[@renee1187]

you just tell about gradient boosting what about extreme gradient boosting ?
tittle is incorrect ....

[@gauravmalik3911]

Best explanation, btw how do we choose learning rate

[@carsten7551]

You can tinker around with the learning rate yourself to see how the model's accuracy improves depending on a larger or smaller learning rate. But keep in mind that very large or small learning rates may not be ideal.

[@firstkaransingh]

Link to xgboost video ?

[@NghiaDuongTrung-k7l]

How about another tree architecture when the root is from another feature? Let's say we start at the root of "is not Blue?"

[@moleculardescriptor]

Something is not right in this lecture. If each subsequent tree is _the_same_, as shown here, then after 10 steps the 0.1 learning rate will be nullified, e.g. equivalent to the scaling = 1.0! In other words, no regularization. Hence, trees must be different, right?

[@charlesmonier7143]

this is not XGBoost. wrong title

[@KalyanAngara]

Dr. Ryan. How can I cite you? I am writing a report and would like to cite your teachings.

[@GeorgeWilliams-v9d]

Harris Carol Harris Edward Jackson Jason

[@davidnassau23]

Please get a better microphone.