Torch No Grad

[ghost]

I am in the process of rewriting my DRL code in Java (from Python) using the PyTorch engine.
Please can I be advised how create a torch.no_grad() context as in;

with torch.no_grad()
etc, etc

Many thanks in anticipation.

[frankfliu]

@BobKelly1212
If you are running inference, by default NDAarray we are not calculate grad.
Would you please provide more detail about what you are trying to achieve.

[saudet]

If you're looking for an API that is close to the Python of PyTorch, you're probably going to be more satisfied with the JavaCPP Presets for PyTorch, which wraps the C++ API of PyTorch, and does have things like "no_grad", see bytedeco/javacpp-presets#623 (comment).