Why use unsqueeze() here?

[mklingamsetty]

In this image, I dont understand why we have to use the unsqueeze method to make the set from a 1D tensor to a 2D tensor. When I ran my code without the unsqueeze method it still worked, the only difference being the output was a 1D tensor.

[Prezzo-K]

It is because most of the time you will work with data with 2 or more dimensions. In that case, it won't make sense if you have your data in one 1D i.e [x1, x2, x3, x4 ...].

But with unsqueeze, the same data is represented as
[
[x1, x2, ...], - sample 1
[x1, x2,...] - sample 2
....
]
and u can add more features to each sample.

Hope this helps :)

[ErsiZhao]

because we want to have them in vectors for later calculations, or else just bunch of number and we cant do forward or backward calculation.