TorchSharp
TorchSharp copied to clipboard
dotnet
A .NET library that provides access to the library that powers PyTorch.
Hello, I am new here, I would like to ask how and when the importance of the draw_bounding_boxes function (https://pytorch.org/vision/main/generated/torchvision.utils.draw_bounding_boxes.html#torchvision.utils.draw_bounding_boxes) will be resolved. Or could someone please advise me how...
Doesnt appear to be any way to transfer a result tensor in to an existing cpu float array. Below requires new memory allocation. var cpuResult = gpuResult.cpu(); float[] result =...
TorchSharp does not support the torch.use_deterministic_algorithms() method. If I try to call it, I get this error: ``` error CS0619: 'torch.use_deterministic_algorithms(bool, bool)' is obsolete: 'not implemented' ``` It would be...
This aligns parameter naming for these functions with the rest of the project. See also: https://github.com/AssetRipper/AssetRipper.Bindings.LibTorchSharp/blob/a76753d26df195f667c003f8f10b3085882168a0/AssetRipper.Bindings.LibTorchSharp.SourceGenerator/ParameterNameChanges.cs#L7-L12
All the other functions with an allocator use size_t. In addition, the `sz` variable in the implementation is `size_t`, and the p/invoke method in `LibTorchSharp.THSJIT.cs` uses the same delegate as...
The native and managed signatures differ. The native implementation has an unnecessary extra parameter.
Reference: https://github.com/dotnet/machinelearning/blob/1c8be6642440cb573a1936dad4b51c465674f1d3/src/Native/build.cmd#L55-L87
A security vulnerability in ImageSharp prevents TorchSharp from building. https://github.com/advisories/GHSA-rxmq-m78w-7wmc
These 3 methods have incorrect signatures for their entrypoint. Each of them has an overload with the correct signature.