-
It would be nice to be able to quickly scrub through a video using the built-in player. I'd suggest the following:
Right/Left Arrow: Forward/Backward 30 seconds
Right/Left Arrow + Shift: Forward/B…
-
Hi there, I'm having OOM errors when running the summarization example on a 80GB A100 (CUDA 11.8).
I'm also getting some Tensorflow/TensorRT warnings, I'm wondering if it's related to that
```
20…
-
For example, if I wanted to augment Gaussian with a variable, I modified the code to include the forward and backward procedures. But when I finally run it, I get an error saying I can't find the defi…
-
help me fix issue @tomaarsen , thanks u guys
```
onnx==1.13
onnxruntime==1.14
pytorch==2.0.0+cuda117
```
![image](https://user-images.githubusercontent.com/77484083/228126260-b0f2f006-da1c-4…
-
╭──────────────────────── Traceback (most recent call last) ─────────────────────────╮
│ /root/llm_riskraider/Baichuan-13B/cli_demo.py:169 in │
│ …
-
### Issue type
Bug
### Have you reproduced the bug with TensorFlow Nightly?
No
### Source
source
### TensorFlow version
tf 2.14.0
### Custom code
Yes
### OS platform and distribution
window…
-
if the selection is like the following (M: mark, P: point)
```
........M.........
....................
....P.........
```
then `meow-line` should turn it into the following and change the se…
-
Hi, great work!!!
How can I look into the implementation of selective_scan_cuda.fwd and bwd?
out, scan_intermediates, out_z = selective_scan_cuda.fwd(
conv1d_out, delta, A…
-
Minimal repro:
```
import torch
m = torch.nn.ReflectionPad2d((1, 2, 3, 4))
m_compiled = torch.compile(m)
inp = torch.randn(2, 3, 8, 8, device='cuda', requires_grad=True)
out = m_compiled(inp…
-
From my point of view it would be a good idea not to query all roads with both maxspeed directions anymore. Otherwise there would be a maxspeed as well as a maxspeed:forward and maxspeed:backward whic…