You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am not sure that it is correct place to ask but Pytorch version 1.0 has the C++ Frontend Pytorch (that allows consuming all variety of neural network modules, optimization algorithms etc. in c++) . Is there easy way to automate c#<->c++ binding to it?
Basically what I am trying achieve is to use high level api of pytorch in C# like
torch::nll_loss torch::log_softmax or torch::nn::Linear etc.
The text was updated successfully, but these errors were encountered:
Hi @artem-hryb, please look here. I will try to submit a PR soon-ish so that you can start playing with it from the main TorchSharp repo. For the moment I am trying to port the API manually so that I can understand if there is any specific blocker. After that we will look on how to automate the process (suggestions are welcome if you already have some idea on how to achieve this!)
I am not sure that it is correct place to ask but Pytorch version 1.0 has the C++ Frontend Pytorch (that allows consuming all variety of neural network modules, optimization algorithms etc. in c++) . Is there easy way to automate c#<->c++ binding to it?
Basically what I am trying achieve is to use high level api of pytorch in C# like
torch::nll_loss torch::log_softmax or torch::nn::Linear etc.
The text was updated successfully, but these errors were encountered: