-
Hey everyone!
First of all, congratulations for your new [Information Retrieval models](https://www.sbert.net/docs/pretrained-models/msmarco-v2.html#performance). They are absolutely amazing.
**My…
-
So, I am taking Biology in high school right now, and I find it really fun. The biochemistry was really interesting to learn about. And then I was playing around with some MineChem. I realized that…
-
Hi, thanks for opening the source code. I read the paper, I find you use logits and features before pooling to perform diffusion. but for the logits, I guess the dimension is [B, C] B is the batch siz…
-
你好，我最近在学习YOLO知识蒸馏算法，我想问下，你的代码里软目标损失计算时为什么除以了batch_size，在最后加上硬目标损失（这部分好像代码没有除以batch_size），而且最后的loss，又乘以了batch_size/64。这部分我看的有点蒙。
能给我解释一下吗？
1、loss_st = criterion_st(nn.functional.log_softmax(output_s…
-
We keep this issue open to collect feature requests from users and hear your voice. Our monthly release plan is also available here.
You can either:
1. Suggest a new feature by leaving a comment…
-
# URL
- https://arxiv.org/abs/2411.19865
# Authors
- Justin Chih-Yao Chen
- Zifeng Wang
- Hamid Palangi
- Rujun Han
- Sayna Ebrahimi
- Long Le
- Vincent Perot
- Swaroop Mishra
- Mohi…
-
This issue is to maintain a list of some algorithms that seem particularly useful to implement. Currently up on the docket are:
- [x] [Deep High-Resolution Representation Learning for Human Pose Es…
-
Hi, thanks for your great work!
It is known that CLIP style contrast loss requires huge global batch sizes (e.g. 32k). I'd like to know if this is a critical issue in your training and the global bat…
-
Would you please release the detailed data of the experimental results on the other two datasets, including miniImageNet and CIFAR100?
-
Hello, I'm Hady an ECE student at cairo university school of engineering, I've been working on a distilled version of a text summarization model called pegasus, I found your L3-AI talk on YouTube and …