-
Right now we have and support "cpu" as a device for all supported libraries in the to_device() helper. In particular, we allow `to_device(cupy_array, "cpu")` to convert a CuPy array to a NumPy array.
…
-
# Issue Description
I found this bug with the python backend while trying to get it working with ROCm for my AMD GPU, though I imagine the same issue may be present for CUDA too. If PyTorch sees a GP…
-
**Description**
When setting dynamic batching and a large batch size in `config.pbtxt` and submitting N individual inference requests, the Triton BLS model iterates through the N requests, executing …
-
https://docs.cupy.dev/en/latest/user_guide/interoperability.html#pytorch
The doc sounds like `__cuda_array_interface__` is preferred over DLPack as it's mentioned first. I'd like to suggest DLPack …
-
### Description
Currently, if you do
```python
cupy_array[some_iterable]
```
CuPy will interrogate `some_iterable` to see if it implements `__array_struct__`, `__array_interface__`, `__array_…
-
Not a bug, but I thought I would make you aware of this pull request from the main JAX devs,
[JAX] Implement importing external dlpack-aware Python arrays, which allows for creating jax.Arrays from…
-
Hello, IREE developers, I've seen a similar question asked before: https://github.com/iree-org/iree/issues/11573#issuecomment-1371128261, but it seems lack of further discussion.
Suppose we have a c…
-
Should DLPack also define a stable format for sparse arrays?
-
### Description
When using `from_dlpack` on an array with non-standard strides, JAX returns the wrong arrays (silently).
Think it's similar to issues mentioned before #4196, #7657, #8082. I under…
-
Hi,
Thanks for your amazing libraries!
I am currently carrying on research using DBSCAN to segment car instance from 3D point cloud. However I find that when i convert pytorch tenosr to DataFram…