What is the difference between torch.Tensor([1,2,3]) and torch.tensor([1,2,3])

[shivammavihs]

I want to understand what is the significance of each function torch.Tensor([1,2,3]) and torch.tensor([1,2,3]).

The one difference I found is torch.Tensor() creates tensors with int64 dtype and torch.tensor() creates float32 dtype by default. Is there any other significant difference between both.

And, when and where to use which one?

[mrdbourke]

Hi @shivammavihs,

I didn't even know they created tensors in different data types!

I'd say I use torch.tensor() far more often.

float32 is a far more often used datatype throughout machine learning/deep learning too.

[kartik-powar-3103]

Hi @shivammavihs

The torch.tensor infers data type automatically whereas torch.Tensor assigns the default data type

I think better explanation can be found in this discussion forum:
https://discuss.pytorch.org/t/difference-between-torch-tensor-and-torch-tensor/30786/2