-
The below example:
```python
import dpnp, dpctl, dpctl.tensor as dpt
dpctl.__version__
# Out: '0.18.0dev0+158.g7450558d25'
a = dpnp.ones(10)
a.flags
# Out:
# C_CONTIGUOUS : True
# F_C…
-
## 🐛 Bug
In the example below, we have 2 tensors: `t0` and `t1`. `t1` is created from a DLPack capsule generated from `t0`. So, we could say they share the same storage. However, after modifying `t…
-
When interoperating with other libraries, `dlpack` has been a useful way to share array data. It might be helpful to have a few utility functions to enable passing `mdarray` in `dlpack` `struct`s or a…
-
**Describe the bug**
When we import pytorch before cugraph we sometimes throw a segfault during cleanup. We should fix that.
**Steps/Code to reproduce bug**
```python
import torch
impo…
-
### 🚀 The feature, motivation and pitch
Pytorch should support using managed memory based dlpack capsules.
### Set Pools
```python3
import rmm
from rmm.allocators.torch import rmm_torch_allo…
-
Hi @tqchen, we've discussed this in a recent array-API meeting and no objection was raised. Given that several projects have successfully migrated to DLPack v1.0, including:
- NumPy (https://github.c…
-
## Motivation
The [`sparse`](https://github.com/pydata/sparse) (also called PyData/Sparse) development team have been working on integration efforts with the ecosystem, most notably with SciPy, sciki…
-
Similar to #10217
Can we run onnxruntime model with cupy array (with some conversions)?
I tried the **dlpack** way as mentioned in #4162 but I got module not found error with the statement `C.Ort…
-
### 🐛 Describe the bug
1. **XPU OP Examples**
57 passed in 58.84s
2. **XPU OP Extended UT**
4 failed, 1233 passed, 309 skipped, 128 deselected, 14 xfailed in 186.95s (0:03:06)
- [x] FAILED te…
-
### 需求描述 Feature Description
Right now, paddle only produce and consume PyCapsules. However, dealing with raw capsules is outdated. Also, numpy doesn't supports PyCapsules. It's better if paddle te…