I have a quick question, but before that I'd like to express my thanks for this wonderful resource. My question pertains to your use of the test data as the validation set. While that is perfectly valid, I was wondering why you don't use the validation split method as well? This would require shuffling of the data prior to avoid biasing the network. But would it not perform an additional layer of proof that your model is not over-fitting? Thanks again for the resources
in 27 line,the result"8s 774us/sample - loss: 0.0141 - accuracy: 0.9913","8s 774us/sample" is what,and "=================================================" is what and,why is 10000/1,what is the meaning,thanks you very much
Hello Sir, You have deleted github all project, I want to download your code and i want to run my local system. Please upload your code github once again, I want to learn Deep Learning follow by your video and code parallel.
I have a quick question, but before that I'd like to express my thanks for this wonderful resource. My question pertains to your use of the test data as the validation set. While that is perfectly valid, I was wondering why you don't use the validation split method as well? This would require shuffling of the data prior to avoid biasing the network. But would it not perform an additional layer of proof that your model is not over-fitting? Thanks again for the resources
Yes that can be used. Instead use k fold cross validation.
sir please provide the code for the same, the github link in the description is not working
thanks for your video but why is reshape(60000,28,28,1)rather than (60000,28,28,3),can you give me an answer,thank you very much
Hi ! the image input is Grey scale, hence only one layer for colour data, 3 would have been the case if it was a colour image.
y it is 28,28,1 what does 1 mean?
Sir i think you might forgot to enter kernal in 2nd conv2d
It is there Akash. I have just not written the name. It is doing same thing.
in 27 line,the result"8s 774us/sample - loss: 0.0141 - accuracy: 0.9913","8s 774us/sample" is what,and "=================================================" is what and,why is 10000/1,what is the meaning,thanks you very much
Hello Sir, You have deleted github all project, I want to download your code and i want to run my local system. Please upload your code github once again, I want to learn Deep Learning follow by your video and code parallel.