-
### Search before asking
- [X] I have searched the YOLOv5 [issues](https://github.com/ultralytics/yolov5/issues) and [discussions](https://github.com/ultralytics/yolov5/discussions) and found no simi…
-
### Contact Details
_No response_
### What happened?
C:\llamafile-0.8.13\bin>llamafile.exe -m \models\Karsh-CAI\Qwen2.5-32B-AGI-Q4_K_M-GGUF\qwen2.5-32b-agi-q4_k_m.gguf -ngl 99 -c 65536 --rope-scali…
-
We are using a scale factor of `8` in the reference implementation, which seems to match with Hugging Face config files for 3.1 models.
However, I observed that the new 3.2 models use a scale facto…
-
### What happened?
When running an experiment in which a model is being replaced by other models via a factor, any disabled children of the replacement models will be enabled before the simulation is…
-
a simple reproducer:
```
import torch
from torch import nn
import sirfshampoo
device = torch.device("cuda:0")
class DebugNet(nn.Module):
def __init__(self, **kwargs):
super()…
-
Hi, I'm training transformer model with Hybrid Sharded Data Parallelism. This setup is similar to FSDP/ZeRO-3 where params all-gather-ed for each layer's forward/backward pass and dropped afterwards. …
-
`trainer.py`'s `train_batch` function has a number of arguments, it would be nice if each of these had an associated type hint, eg `bool`, `int`, `EmaModel`, etc.
```python
def train_batch(
m…
-
Hi, great work! I have questions about this models
1. I was initially planning to use abodybuilder2 for nanobody structure generation (Nanobodybuilder).
Would it be approapriate for me to use ab…
-
Example code:
```py
import tensorflow as tf
X = tf.keras.layers.Input(shape=(60, 78),
name='factor')
x = X
y = tf.keras.layers.GRU(1,
…
-
The initial value and gradient is undefined because the start value is `y = 0` and there is `1 / y^2`
This could be classified as expected behavior?
Ipopt is able to solve the model.
## Uno
…