-
Is there any value in supplying a pure_callback (one with a return value), a buffer into which to place it's output? Something like the code below. I have some science applications where I need to use…
-
Requested in pytorch/ao#292 and pytorch/pytorch#38349 (NumPy rollup).
cc @mruberry @rgommers
aayn updated
2 months ago
-
The [array API specification stipulates](https://data-apis.org/array-api/latest/API_specification/array_object.html#size) the following operators that are currently not supported by PyTorch:
- [x] …
-
A related issue in onnxruntime: https://github.com/microsoft/onnxruntime/issues/4162. Not sure where this feature request should belong
There is [DenseTensor](https://github.com/microsoft/onnxrunti…
-
你好，參考您的blog文章，遇到`Tensor is stored in GPU and cannot be converted to NumPy.`，使用torch的`from_dlpack`接口進行轉換。
發現首次執行時，會耗時非常久(15分鐘)，之後就會比較正常(
k0286 updated
2 years ago
-
### Please describe your wishes and possible alternatives to achieve the desired result.
With the [array-api](https://data-apis.org/array-api/latest/) it should be possible to have some level of sup…
-
```
In file included from /usr/ports/misc/tvm/work/tvm-0.11.0/src/arith/canonical_simplify.cc:24:
In file included from /usr/ports/misc/tvm/work/tvm-0.11.0/include/tvm/arith/analyzer.h:27:
In file …
-
**Is your feature request related to a problem? Please describe.**
It is not currently straightforward to pass external dataloaders to train a model. In particular, loading `torch.Tensor` data and di…
j-bac updated
3 months ago
-
For dataframe interchange, the smallest building block is a "buffer" (see gh-35, gh-38) - a block of memory. Interpreting that is nontrivial, especially if the goal is to build an interchange protocol…
-
### Problem description
An extension that returns an `ndarray` will give the user the following DeprecationWarning:
> Calling `from_dlpack` with a DLPack tensor is deprecated. The argument to `fro…