-
Currently the `library-image` addon makes use of a `BufferedImage` wrapper to manage images. This works fine for relatively small images but has issues for very large images that on disk may be very s…
-
### 🚀 The feature, motivation and pitch
We have many tensors with large shape, but several stride=0 dimensions, i.e. expanded/broadcasted dimensions.
If a reshape that cannot be done as a view is …
-
Representing `BooleanSeries` internally as a bitset instead of `boolean[]` results in a very compact data structure (1/8 of byte vs 1 byte). Access is probably slower , but we can improve it by replac…
-
raise NotImplementedError(msg)
NotImplementedError: No operator found for `memory_efficient_attention_forward` with inputs:
query : shape=(40, 1564, 1, 64) (torch.float32)
key …
-
No operator found for `memory_efficient_attention_forward` with inputs:
query : shape=(1, 577, 16, 64) (torch.bfloat16)
key : shape=(1, 577, 16, 64) (torch.bfloat16)
value : shape=(1, 577, 16, 64) …
-
@mweidner037 It was great meeting yesterday! After our talk, I've been thinking about two things you said: (a) that translating the position to the run length encoding is the trickiest part of the cod…
-
https://github.com/yukiregista/booster/issues/16 のタスクが終わったら、Python側も対応する。
-
Run the following code on a GPU with python:3.10 CUDA:12.4 torch:2.4.0
```python
import torch
from xformers import ops as xops
from xformers.ops.fmha.attn_bias import BlockDiagonalMask
torch.…
-
# ❓ Questions and Help
Currently, I have successfully built PyTorch 2.5.1 from source (with CUDA 12.6.3 and cuDNN 9.5.1) and installed it into a Miniconda3 Python 3.11.10 virtual environment. I also …
-
### Describe the bug
> RuntimeError: The size of tensor a (154) must match the size of tensor b (2304) at non-singleton dimension 1
### Reproduction
```python
# StableDiffusion3Pipeline
pipe.enab…