I am so glad to have come across this course, thank you for teaching the way you do, please I have a question, do we have to memorize all this codes you use to use them? or there are they written somewhere in the python libraries? thank you.
I have taken part in WIDS Datathon where the data is too big...so please can u suggest me how to handle the missing values..the problem is Independent And Dependent variables are very difficult to handle
Thank you so much. it helped me. but now i have a large dataset with lots of missing value. i think it's not good to remove that rows. what solution you suggest?
Try to replace with mean. If your missing values column just have few unique values the try to replace with mod. What would be the best choice it all depends on what data is, how many values are missing, meaning of that value and unique values in that column.
Try with mean and mod and see what works best. I would also recommend to drop these 4000 records and see what's the results you are getting with model.Then select the one which work better for your business problem.
For deep learning data preprocessing will use in same concept or not. And for image type dataset how data preprocessing will do. Pls kindly reply for my doubt
If your data is intabular form then similar data preprocessing is need irrespective of ML or Deep Learning. If you are using image data the data preprocessing will be different checkout our computer vision with openCV playlist. If you data is just plain text in that case text preprocessing will be different for NLP projects.
I am so glad to have come across this course, thank you for teaching the way you do, please I have a question, do we have to memorize all this codes you use to use them? or there are they written somewhere in the python libraries? thank you.
Is it necessary to replace missing data ? Cant we work on the data after droppping all missing data?
I have taken part in WIDS Datathon where the data is too big...so please can u suggest me how to handle the missing values..the problem is Independent And Dependent variables are very difficult to handle
Thank you so much. it helped me. but now i have a large dataset with lots of missing value. i think it's not good to remove that rows. what solution you suggest?
Try to replace with mean. If your missing values column just have few unique values the try to replace with mod. What would be the best choice it all depends on what data is, how many values are missing, meaning of that value and unique values in that column.
@@DataMagicAI number of rows is 60000 and rows without missing value is just 4000. doesn't replacing with mean or mod change the actual result?
Try with mean and mod and see what works best. I would also recommend to drop these 4000 records and see what's the results you are getting with model.Then select the one which work better for your business problem.
Are u working on train or test data set
what if the missing values are object type then what do we do?
very helping videos, Thanks a lot 😊
Most welcome!
For deep learning data preprocessing will use in same concept or not. And for image type dataset how data preprocessing will do. Pls kindly reply for my doubt
If your data is intabular form then similar data preprocessing is need irrespective of ML or Deep Learning.
If you are using image data the data preprocessing will be different checkout our computer vision with openCV playlist.
If you data is just plain text in that case text preprocessing will be different for NLP projects.
@@DataMagicAI thank you so much
really helpful
Glad to hear that
Very helpful Video 👏 Keep it up
Sure ! Thank you !
inplace=True is not for the purpose that you mentioned.
very helpful thanx
Glad it helped
Glad it helped
In my college they confused me with these topics .... but u made these topics very easy for me to understand ..... tq u sm sir
Sir, how to fill the missing data with zero ?
dataframe.fillna(0)