gchanan
Follow
Create your own GitHub profile
Sign up for your own profile on GitHub, the best place to host code, manage projects, and build software alongside 50 million developers.
Sign up
Popular repositories
-
-
Forked from pytorch/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
-
-
-
1,025 contributions in the last year
Contribution activity
August 2020
Created a pull request in pytorch/pytorch that received 3 comments
Fix CUDA mode strided output and add generic test for strided multi outputs.
Stack from ghstack: #43152 Fix CUDA mode strided output and add generic test for strided multi outputs. Differential Revision: D23170685
+71
−3
•
3
comments
- Adds arccos, arcsin, arctan aliases
- Add a composite function torch.fast_conj()
- Enables bfloat16 x [float16, complex64, complex128] type promotion
- Fix torch.hub for new zipfile format.
- Add torch.dot for complex tensors
- Optimize SiLU (Swish) op in PyTorch
- Replace all AT_ASSERTM under ATen CUDA kernels.
- fix searchsorted output type
- Change C++ frontend to take optional<Tensor> arguments
- Add _foreach_add_(TensorList tensors, Scalar scalar) API
- Add _foreach_add(TensorList tl1, TensorList tl2) and _foreach_add_(TensorList tl1, TensorList tl2) APIs
- Don't materialize output grads
- Adds torch.linalg namespace
- CUDA reduction: allow outputs to have different strides
- Adds fft namespace
- Add private API to support tensor lists: _foreach_add(TensorList tensors, Scalar scalar)
- [JIT] Add JIT support for torch.no_grad
- Add missing space after -> for topk.values
Created an issue in pytorch/pytorch that received 1 comment
Add a guide for what to do when you think there's a memory leak
1
comment