Powering AWS purpose-built machine learning chips. Blazing fast and cost effective, natively integrated into PyTorch and TensorFlow and integrated with your favorite AWS services
Is there a way to automatically derive the bitwidth given a type? For example, if I have a tensor A, I can get its type with A.dtype, but can I get the bitwidth of the individual elements? In Numpy, we have itemwidth (see here). Thanks!
Is there a way to automatically derive the bitwidth given a type? For example, if I have a tensor
A
, I can get its type withA.dtype
, but can I get the bitwidth of the individual elements? In Numpy, we haveitemwidth
(see here). Thanks!