-
# 🚀 Feature request
Is it possible to sample the negatives for the Two-Tower model from a column provided by the input data?
For example, we want to sample negatives from the list of items we disp…
-
can squeezenet be used for speech emotion recognition if we feed 3D log mel spectrum values?
-
To the Authors
This is a very interesting and good work on visual grounding tasks with a Query-based detector. The paper is also well written and clear. Super interesting results with GLIGEN as we…
-
Fantastic work, John. I greatly enjoy exploring the package thus far.
I've run into some problem in estimating a WTP-space model. The problem appears when I requested a larger number of draws. The …
-
### Describe the bug
When using automatic mixed precision with either `float16`, `bfloat16`, I encounter a casting issue during the **backward** pass:
```plaintext
Traceback (most recent call l…
-
Thanks for the awesome package. Cristian.
For implementation of WTP Space models, having reasonable starting values may be critical. My attempt to estimate my own data resulted in convergence issue…
-
Hi Joey,
Thank you for such a wonderful OS work! !
Could you share the exact command to reproduce the curve in your MOD is Vibe blog? For example, did you use DDP and how many GPUs?
-
I see people are trying to extract the Mistral-22b ancestor from the MoE model by averaging the MLP layers and wondered if the 'model stock' method in Mergekit could be inverted:
- Use the averaged…
-
我尝试给yolo_world_v2_l_vlpan_bn_2e-4_80e_8gpus_mask-refine_finetune_coco.py中直接添加
mg_train_dataset = dict(type='YOLOv5MixedGroundingDataset',
data_root='data/mixed_grounding/',
…
-
We are trying to run the [e5 model](https://huggingface.co/intfloat/e5-large-v2) on an inf2 instance. The model compiles fine and analyze reports no unsupported operators but when trying it out on an …
aabbi updated
3 months ago