Deep Learning with Class Imbalance in R Notebook | Using Keras and TensorFlow

[@amirgharavi8820]

Thank you so much for the video and explanation Dr.Rai. Great as always

[@bkrai]

Glad it was helpful!

[@amirgharavi8820]

@@bkrai Fantastic Dr Rai. very much appreciate it.

[@shwetaredkar734]

Thank you very much for such helpful demonstration.

[@bkrai]

Thanks for comments!

[@liuxinzhu89]

Great videos! Using your video I managed to achieve 86% in the first learning!

[@bkrai]

That’s great!

[@flamboyantperson5936]

Happy New Year Sir. I have been long waiting for your videos. Thank you so much for making tutorial on Deep Learning. Great Knowledge.

[@bkrai]

Thanks for your comments!

[@praveengupta3918]

Hi Sir, Your videos are helping me from my college days. Thank you

[@bkrai]

Thanks for your feedback!

[@ravindarmadishetty736]

Happy new year sir....good video in new year

[@bkrai]

Thanks for comments!

[@ravindarmadishetty736]

Sir, have a question I have to get the data from Hive using R. Will it possible to get the data using r from Hive.... please suggest

[@asjsingh]

Hello. Another great tutorial. I seem to have a problem of no class_weights function in my keras. It doesn't pop up as an option and gives an error when I run it. Do you know if the ckass_weights function has been depreciated? I can find no documentation to this effect.

[@doitonath]

Hello Dr. Rai. Can you please make a video on hypertuning of various parameters of DL in Keras R (like the same way we have in H2O R)

[@bkrai]

Thanks for the suggestion, I've added it to my list.

[@Didanihaaaa]

Thanks, Dr. Could you teach us how to use Atom for running R codes?

[@arsalanriaz7784]

Thank you for the tutorial, it is just brilliant. what if we set loss equal to focal loss instead of direct cross-entropy, how you will consider this change?

[@bkrai]

You can run this and see if results improve.

[@rgemsph7339]

Hello Sir Thank you for this tutorial it really helped me alot but I have a question, may i ask how to know the optimal batch size and epoch when you have large dataset? in my case I have 150000 observation

[@bkrai]

It depends on the data. You may have to do some trials. See this link as an example: kzbin.info/www/bejne/Y4ScqqOdhs6ifKc

[@redarabie7098]

Thank you Mr Rai for your useful tutorial. it's very useful for us and it helps us to improve our comprehension for deep learning . i have a question about the number of the hiden layer or layer dense , howe can i chouse the optimum number in a study or ho can i creat a loop to test many layer in the same time ? and thank you very much

[@ramp2011]

Great video. Thank you for the share. NZeroes looks like a categorical variable. Right? Should one not do the same to_categorical() for that? Its very good to see a Deep Learning video in R. These days most examples I see are in python for deep learning. Not sure why.

[@bkrai]

Thanks for comments! Note that NZeros is not a categorical variable.

[@rishabhojha8798]

Hey Bharatendra, what does that mean adding 8 to 168 as bias term and adding 3 to 24 as bias term. Please make me understand. Thanks

[@bkrai]

I would recommend going through following from the deep learning playlist for detailed calculations and meaning: Neural Networks in R: Example with Categorical Response at Two Levels kzbin.info/www/bejne/Y4fWaomXmpd-f5I

[@azmatraja5767]

a

[@bkrai]

Here is a way to overcome such errors: kzbin.info/www/bejne/jofJn2l_pdSIhrc

[@adityapatnaik7078]

can you please make a video on how to make an webapp for generating ML predictions

[@bkrai]

Thanks, I've added this to my list.

[@ramp2011]

Can you please share the R file?

[@bkrai]

I've uploaded the link now.

[@ramp2011]

@@bkrai Awesome. Thank you

[@charanlobo2981]

sir...for 3 class variables it works fine...but for 4 class variables... it doesnt work...even the confusion matrix doesnt show the 4th class variable....can u plzz help me with this problem....thanku

[@bkrai]

Note that in the last layer I used units-3 because I had 3 classes. If you have 4, you should use 4.

[@yogitasolanki4638]

Sir I want your help

[@bkrai]

let me know where I can help.

[@yogitasolanki4638]

How can contact u

[@bkrai]

you can use this or send me an email.

[@yogitasolanki4638]

Sir your mail id ?

[@yogitasolanki4638]

I am unable to install_keras()

[@bkrai]

Are you using Mac or Windows?

[@yogitasolanki4638]

Window

[@djdekabaruah3457]

@@yogitasolanki4638 Hi.. i am also facing the same issue for installing keras..Were you able to resolve the problem?