-
### Description
I have an editable local installation in my project directory. It contains a `.so` file build in AMD64.
While building a 32-bit wheel in the docker container with:
```console
$…
-
### Required prerequisites
- [X] I have read the documentation .
- [X] I have searched the [Issue Tracker](https://github.com/metaopt/optree/issues) that this hasn't already been reported. (comment t…
-
PyTree manipulation is now widely used in PyTorch. It becomes a fairly significant perf bottleneck during `torch.fx` tracing using the pure-Python approach of PyTree. We are pushing to use C++-based P…
-
The test will break if I change the assignment to an `as` import.
```python
# SUCCESS
from optree import PyTreeSpec
TreeSpec = PyTreeSpec
# FAILURE
from optree import PyTreeSpec as TreeSpec
…
-
The bloq example "qubitization_qpe_sparse_chem" does not decompose all the way down. Minimal example is
```python
bloq.as_composite_bloq().flatten()
```
```
ValueError …
-
Currently, when there are two device meshes (`tp` and `dp`), torchtitan should choose FSDP as the **only** backend for DP. Ref:
https://github.com/pytorch/torchtitan/blob/d2a4904f58accc683c17c66a3600…
-
## Motivation
OpTree uses a generic custom class that returns a `Union` alias in the `__class_getitem__` method.
For example:
```python
from optree.typing import PyTree
TreeOfInts = PyTre…
-
### 🐛 Describe the bug
Reproducing example:
```
>>> import torch
>>> from torch.fx.experimental.shape_inference.infer_shape import infer_shape
Traceback (most recent call last):
File "", lin…
-
### Required prerequisites
- [X] Make sure you've read the [documentation](https://pybind11.readthedocs.io). Your issue may be addressed there.
- [X] Search the [issue tracker](https://github.com/…
-
### 🐛 Describe the bug
When running the following code:
```
import torch
import numpy as np
x = torch.tensor(-np.inf)
y = torch.tensor(0.9)
res1 = torch.nn.functional.binary_cross_entropy_wit…