Help needed in training my ANN

#1

While training my Artificial Neural Network, the loss function is not attaining any numeric value rather it is showing nan for all epochs, can you help?

#2

hey @anshufirefox ,
you need to check your loss function.

Its calculation is resulting in a nan value , this can be due to

  1. divide by zero being happening
  2. wrong gradient descent rule
  3. weights are not being updated in correct way.

These are some common things , we got previously. You can check that , and see if it helps.

#3

How can I share the code and dataset with you?

#4

you can upload it on github and share me the links

#5

https://colab.research.google.com/drive/1_wn0qzPGlh5cOuLb60tTAAsh-ME-dnLm?usp=sharing

#6

Here, I’ve shared a google colab notebook, hoping that it is convenient

#7

https://drive.google.com/file/d/1ZYf84wAUXNgUXNujlxBGBLqpK4ruGJM7/view?usp=sharing

Here is the dataset

#8

hey @anshufirefox ,
the data that you are using is , contains nan values , try once removing them.

#9

Can you be more specific please, which column or which feature contains nan values?

#10

in training X data , check the last record
and in the target data , there is also some values that are nan.

#11

I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.

On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.