torch.tensor、torch.as_tensor do not always use the current device for the default tensor type when device param is none · Issue #62146 · pytorch/pytorch · GitHub
feature request] torch.as_tensor to support any object that NumPy's asarray or array can consume (consume __array_interface__) · Issue #58036 · pytorch/pytorch · GitHub
Optimize PyTorch Performance for Speed and Memory Efficiency (2022) | by Jack Chih-Hsu Lin | Towards Data Science
01 Tensor Operations - Notebook by Jake Johnson (xyclos) | Jovian