mikey_t

Hi, I have a similar question about runtime. I'm deploying a model in R that I trained in Pytorch and it's taking a bit longer to deploy in R (on...

Great! Thanks a lot for your input! I didn't realize there was a `torch_softmax` function so I'll incorporate that too. I really appreciate your quick responses!

@ccordoba12 Here is what it looks like in my RStudio. To get the code to run in the terminal in RStudio, I press a combination of [Command + alt +...

I would find it useful if when I exported annotations I could get a list of the image name (the file path) and the image number (2) in the example...

Thank you @dfalbel . We'll get working on it. I trained the model in python, but I'm trying to make it deployable in R as I'm making an R package...

Thank you @dfalbel ! This looks like a great option. It would be a huge advantage to not have to port the code to R. I have tried this, but...

Yes-I can understand why this will be tricky. We'll be working on the same problem on our end as we need to include this in an R package. Please let...

Thank you. This looks like a great option. I have not been able to build and create that file. Would you be willing to share the `libtorchvision.dylib` file that you...

I'm having a new problem when I try to load a state_dict from python into R. I'm not having problems with the .pt file that I created above. I'm wondering...

Thank you! My mistake. I was trying to load the architecture as if it was the weights.