-
### Motivation
want to improve the image copy rate from CPU to GPU, how to use the feature defined in th dlpact.h
/*! * \brief Pinned CUDA CPU memory by cudaMallocHost */ kDLCUDAHost = 3…
-
Hey,
Recently Tensorflow and Pytorch have implemented a way to pass tensors from one framework to another using DLpack, and I was wondering if it is possible to somehow use your code in a pytorch mod…
-
## 🐛 Bug
When converting a DLPack object representing an empty CUDA tensor to a PyTorch tensor, it crashes with "CUDA error: invalid argument".
## To Reproduce
Steps to reproduce the beha…
-
### Describe the bug
For large payloads BentoML's Numpy Protobuf serialization/deserialization is ~1000x slower and the JSON serialization/deserialization is ~3000x slower compared to either:
1.…
-
### Describe the feature request
Currently, the DLPack protocol can only be used in non-training build. See:
- https://github.com/microsoft/onnxruntime/issues/4162
- https://github.com/microsof…
-
This is a tracking issue for 2023.12 support in array-api-compat.
The list of changes for 2023.12 is at https://data-apis.org/array-api/latest/changelog.html#v2023-12
Currently, we don't have te…
-
Hey, this is not so much an issue as a place to solicit public feedback.
I think we should implement type conversion from the amrex FArrayBox (or more precisely the Array4) data type to numpy.ndarray…
-
Related to #10754, the current implementation of `from_dlpack` requires unit-stride fortran order, and produces appropriate error messages in the unsupported cases
Consider
```python
import cud…
-
Sooo DLPack is a great, cheap, and flexible data structure to pass between the nodes but tensor shapes aren't a great way to identify what the data actually is. Maybe we have some sort of GUID structu…
-
Is it possible to share dl_pack tensor across python process similar to `multiprocessing.shared_memory`? If possible any documentations on the same? Thanks