Machine Learning Project | Credit Risk Analysis | Learning Curves | Overfitting | Python
#machinelearning #python #project
In this video we will look at a Machine Learning project that will try to predict whether someone will get their loan sanctioned or not.
We will use a Randomized Search too find optimal set of parameters. We will then use precision recall curves and learning curves to assess the model performance.
We will rectify the case of overfitting in the model and make amends to make to more robust and generalized.
Link to dataset: www.kaggle.com/laotse/credit-...
Github link for the project:
github.com/rachittoshniwal/ML...
Thanks!
Пікірлер: 13
thank you so much , very nice explanation
Awesome explanation. I learned a lot, for me learning curve was new. Thank you for such wonderful tutorial
@mathematics-in3wi
8 ай бұрын
Sir Is it good decision to work in the field of finance by using neural network ?
Hi Rachit! Once again, your content is more than great! Why did we drop the "loan_percent_income", since it has the combined information of the 2 other variables? Wouldn't it help to decrease the dimensions of our X matrix? Since we are not going to use the interaction of person_income*loan_amount in our ML model, wouldn't it be better if we use 1 variable instead of 2? Thank you and please keep those tutorials coming. You are a gifted teacher
Great video, Rachit. Thank you. I also heard OneHotEncoding is not good for large categorical data in real world projects. Please which method do you advise or is there a video of you doing it that we can watch? Thank you so much
do you told me sir from where the data set is taken?then i might be thankful
Nice video. I believe rather than deleting the records, one should try to replace those missing / absurd values with average values in the column.
Hello Rachit I like your approach to solve Machine Learning Problems but i am not able to map those steps to another ML problems (Eg: Kaggle Competitions ) could you make a video on how approach ML problems?? easily like you, It would be very helpful. :)
@rachittoshniwal
3 жыл бұрын
How to approach ML problems as in?
@abhishekgadde5668
3 жыл бұрын
@@rachittoshniwal Like kaggle challenges or topcoder challenges they are little bit tougher
@rachittoshniwal
3 жыл бұрын
@@abhishekgadde5668 I see, currently I'm trying to make beginner to intermediate level stuff to get the ball rolling, you know. I'll look into it though, thanks!
Where data is available
@rachittoshniwal
2 жыл бұрын
It is on my GitHub