Asume u r in class 12 and preparing for JEE Mains. U are an excellent student of Mathematics and Physics but u barely understand chemistry. If Boosting algorithm is ur teacher, it ensures that u get more class of chemistry so that u focus on the areas where u are weak. This is exactly what boosting does in a laym man terms. It ensures that u see more samples of data which u are predicting badly so that ur performance in those areas increase and thus u get a better score. It also has some regularisation effect inbuild thus reducing over fitting also to some extent.
Honest feedback for Arslan and many such freshers - Kindly focus and work on your communication skill as well. Many of you will get rejected if you are not able to communicate well
The way sudhanshu sir used for asking question is awesome😊 Great job sudhanshu sir and krish sir.These interview series help us to how to prepare for Data Science interview.
For those who are learning about boosting, here's the crux. In boosting, we first build high bias, low variance (underfitting) models on our dataset, then we compute the error of this model with respect to the output. Now, the second model that we build should approximate the error that we have for our first model. second_model = first_model + (optimisation: find a model which minimises the error that the first model makes) This methodology works because as we keep on building the model the error get's minimised, hence the bias reduces. So, we get a robust model. Going a bit more in depth, instead of computating the error we compute the pseudo residual because the pseudo residual is proportional to the error, and we can minimise any loss. So, the model becomes, model_m = model_at_(m-1) + learning_rate * [derivative of the loss function with respect to model_at_(m-1)]
There is no such term weak learner in whole boosting.. We only use the term weak learning in adaboost.. Because it creates a stumps instead of trees. Stump is like when you have an decision tree with root and two or more leaves.. In that time it can't able to predict very well.. So we will combine those weak learners to make it strong learners
sir i want to pursue my carrier in data analytics, i have most of the videos, and also tried some solve problem . But when i am on my way to do analysis on any data set, i get stuck after doing describe funtion, i am not able how to understand read and analyse the output and move forwad. please help me any way, or suggest me some videos to follow for better clarification..
Sir please please 🙏 explain which course is better for making career into data science like MCA/PGDBA/Msc data science It will be very very helpful for students like us who had done their graduation in bsc IT/BCA/bsc Statistics etc.... And genuinely want to make their career into Data Science!! Sir please, please make a video on this topic Sir could you please make a video again on Top PGDBA or Mba in business Analytics colleges in india and top Msc Data science/Big Data Analytics colleges in India
Asume u r in class 12 and preparing for JEE Mains. U are an excellent student of Mathematics and Physics but u barely understand chemistry. If Boosting algorithm is ur teacher, it ensures that u get more class of chemistry so that u focus on the areas where u are weak. This is exactly what boosting does in a laym man terms. It ensures that u see more samples of data which u are predicting badly so that ur performance in those areas increase and thus u get a better score. It also has some regularisation effect inbuild thus reducing over fitting also to some extent.
Excellent explanation!
Just getting good information. Nothing else. I attending interview passively. This is my fourth Interview 😂 passively. ❤️
Honest feedback for Arslan and many such freshers - Kindly focus and work on your communication skill as well. Many of you will get rejected if you are not able to communicate well
The way sudhanshu sir used for asking question is awesome😊
Great job sudhanshu sir and krish sir.These interview series help us to how to prepare for Data Science interview.
For those who are learning about boosting, here's the crux.
In boosting, we first build high bias, low variance (underfitting) models on our dataset, then we compute the error of this model with respect to the output. Now, the second model that we build should approximate the error that we have for our first model.
second_model = first_model + (optimisation: find a model which minimises the error that the first model makes)
This methodology works because as we keep on building the model the error get's minimised, hence the bias reduces. So, we get a robust model.
Going a bit more in depth, instead of computating the error we compute the pseudo residual because the pseudo residual is proportional to the error, and we can minimise any loss.
So, the model becomes,
model_m = model_at_(m-1) + learning_rate * [derivative of the loss function with respect to model_at_(m-1)]
Sir you both are doing great love u...
I will join full stack ds course this time. And recommend my friend Arslan Eqbal Khan to take it
That's why derivations are more important.
Yep many ppl just use without knowing how it works
@@omyerawar7976 that's why i recommend to take your time and learn the algorithms thoroughly rather than just rushing through.
There is no such term weak learner in whole boosting.. We only use the term weak learning in adaboost.. Because it creates a stumps instead of trees. Stump is like when you have an decision tree with root and two or more leaves.. In that time it can't able to predict very well.. So we will combine those weak learners to make it strong learners
Thank you sudanshu sir and Krish Naik sir
Hope sir all the interviewers are like Sudhanshu sir.....
Great session Sir!!
Nice
sir can u share candidates resume also this will help us in how to make resume as a fresher and what should we mentioned in resume for shortlisting.
good ask
sir i want to pursue my carrier in data analytics, i have most of the videos, and also tried some solve problem . But when i am on my way to do analysis on any data set, i get stuck after doing describe funtion, i am not able how to understand read and analyse the output and move forwad. please help me any way, or suggest me some videos to follow for better clarification..
Sudhanshu sir and Krish sir having fun interviewing arslan😂
Sir please please 🙏 explain which course is better for making career into data science like MCA/PGDBA/Msc data science
It will be very very helpful for students like us who had done their graduation in bsc IT/BCA/bsc Statistics etc....
And genuinely want to make their career into Data Science!!
Sir please, please make a video on this topic
Sir could you please make a video again on Top PGDBA or Mba in business Analytics colleges in india and top Msc Data science/Big Data Analytics colleges in India
Good entertainment 😆
Interviews of data/business analysts?
He seems to be reading all this concepts somewhere
Sudhanshu like i want story .. story nikaal bhen***
DL concepts only
Bhai arslan bahut Shi jaa RHA h tu
Good Arslan ❤️