tch-rs icon indicating copy to clipboard operation
tch-rs copied to clipboard

Rust bindings for the C++ api of PyTorch.

Results 213 tch-rs issues
Sort by recently updated
recently updated
newest added

Thanks for the hard work on this project! The current installation instructions can be error prone as the torch developers do not expose the 1.12.0 download link readily, and the...

I'm trying to use the tch crate for a rust project on an M1 Mac. I've successfully compiled libtorch following [this](https://github.com/pytorch/pytorch/blob/master/docs/libtorch.rst) tutorial. However, when I try to run a simple...

Hello! I found little bug in `build.rs` and would like to help :D ## Problem NixOS distributive (and some more) separating `libtorch` shared symbols and headers. And it would be...

The URL [https://download.pytorch.org/libtorch/cu111/libtorch-shared-with-deps-1.11.0%2Bcu111.zip](https://download.pytorch.org/libtorch/cu111/libtorch-shared-with-deps-1.11.0%2Bcu111.zip) is AccessDenied now. An upgrade to `cu113` is necessary.

I have an M1 Mac running Monterrey 12.1. I installed pytorch via: ``` $ conda create --name tensorflow_m1 python==3.9 $ conda activate tensorflow_m1 $ conda install -c apple tensorflow-deps $...

I noticed that there isn't a way to easily compare tensors for equality, so I added a macro. I also added a bunch of tests that uses the macro. Let...

I noticed that InstanceNorm was not implemented. I tried to follow the same pattern as BatchNorm with a few exceptions. For BatchNorm, the current implementation always uses weights and biases...

On win10, torch 1.11 and tch-rs 0.7.2: ```rust let tensor_to_transform = Tensor::try_from(array![ [0.2f32, 1.0, 0.0], [0.0, 0.9, -0.14], [0.0, -0.2, 1.0] ]) .unwrap(); let mut tensor_to_transform_2 = Tensor::try_from(array![ [0.2f32, 1.0,...

This API could be found in 0.7, but not 0.8. Why delete it? Is there some replaced API for it?

Hello tch-rs team, Is the transformer in nn model available? I cannot find it in the src directory. It could be very relevant to a lot of fields. Thanks, Jianshu