Hi, many thanks for this awesome repo. I have been trying multiple methods to get TensorFlow to work in UE4 and it is a nightmare, so I'm grateful for this repo.

I'm asking if you could please supply a quick tutorial for using SimplePyTorch now that I have successfully integrated the plugin into my project.

How can I use SimplePyTorch for inference? For example:

1. Uploading my model from a file
2. Making a prediction using the loaded model

If I can then get it working I'd love to create a YouTube tutorial for SimplePyTorch to help out others and increase interest in this very useful repo. It would be great if I could fulfil my main aim of creating a CNN inside UE4. Many thanks.

Note that the above link has CPU-only libtorch. If you would like to download a GPU-enabled libtorch, find the right link in the link selector on https://pytorch.org

is this compatible with GPU lib or should i download only the CPU ?

Thank you for your Plugin,it helped me a lot !

bool Create(TSet<int32> TensorDimensions);

When I input like this InputData.Create({1,10,10,10}); Final TSet equals to {1,10}

Thanks for the tool! Very useful.

Would it be a difficult task for me to expand the list of functions available? For example, to include functions to create and train the model? Is it simply a matter of following the methodology you used for the other functions (e.g. to load the existing script)?

I have tried opening the wrapper dll in Dependency Walker and Visual Studio but it does not seem possible to edit it. Would you be able to recommend any program to view the wrapper dll and add to the functions?

Thanks in advance. I am very new to programming and ue4, so any advice would be greatly appreciated.