Tensor reshaping without corrupting data

[Topsoil]

I’ve noticed that different methods of reshaping a tensor (.reshape(), .view(), .permute()) seem to sometimes somehow corrupt (somehow change its order) data inside tensor, but i was not able to pinpoint when they do so, because reshaping it back restores original

Can someone explain to me what is a preferable method, and in what case i should use each method, please?

[JuanFMontesinos (Juan Montesinos)]

Difference between view, reshape and permute

reshape tries to return a view if possible, otherwise copies to data to a contiguous tensor and returns the view on it. From the docs: Returns a tensor with the same data and number of elements as input , but with the specified shape. When possible, the returned tensor will be a view of input . Otherwise, it will be a copy. Contiguous inputs and inputs with compatible strides can be reshaped without copying, but you should not depend on the copying vs. viewing behavior. See torch.Tensor.view…

Difference between 2 reshaping operations (reshape vs permute)

Im not sure about reshape. Permute is a multidimensional rotation saying somehow. It keeps the data ordering. View (which is another reshaping method) maps from one dimensionality to another sequentially reading data from the upper dimensions to the lower ones. So if you want fuse two dimensions into one, you have to apply it over contiguous dimensions or u will modify the data ordering