Why are we making them tensors when at the end we have to convert them to numpy arrays only? I am not able to understand the concept of tensors like why do we need them?
Tensors vs Numpy
hey @rishitagoel13 ,
Tensors and numpy arrays are almost same , just difference is when we need to use and perform operations.
SO , if you use keras, then you can use numpy as it works absolutely fine.
But , when we are implementing are code completely in tensorflow , then we need to change them to tensors, the tensorflow understand them fast , You can provide numpy there too , but that would be a bit slow. and tensor operations are bit different from numpy , but most of them are same.
I hope this helps.