-
- https://arxiv.org/abs/2107.13034
- 2021
機械学習アルゴリズムの有効性は、大量のデータから有用な特徴を抽出できるかどうかで決まる。
モデルやデータセットのサイズが大きくなるにつれて、大規模なデータセットを大幅に小さくしながらも性能の高いデータセットに圧縮するデータセット蒸留法は、学習効率や有用な特徴抽出の点で価値が高くなる。
この目的のために、…
e4exp updated
3 years ago
-
### Feature request
I am currently working on a project that involves sequence level distillation across multiple domains, requiring the handling of separate datasets for each domain within a single …
-
Hi,
I noticed that you trained the NLP emulator with the first 30 chunks of Pile dataset. I wonder how large are the 30 chunks? Or in other words, how many chunks does Pile have? The original Pile …
-
Hi, I am trying to replicate and pretrain BLIP for distillation purposes - I am using Flickr30K + COCO and my ITM loss gets stuck at 0.63 - upon an initial look, all of the ITM predictions are 1. Is t…
-
### Search before asking
- [X] I have searched the Autodistill [issues](https://github.com/autodistill/autodistill/issues) and found no similar bug report.
### Bug
Hi
I am following this tutor…
-
## 🐛 Bug
According to the [doc](https://github.com/pytorch/fairseq/tree/master/examples/nonautoregressive_translation#knowledge-distillation), it said "The easiest way of performing distillatio…
-
The error is following:
File "D:\PyCharm_workspace\KD\Knowledge-Distillation-via-ND-main\CIFAR\ReviewKD++\utils.py", line 62, in project_center
loss += 1 - torch.dot(s, e_c) / max_norm
RuntimeE…
-
Thank you for your outstanding work in the field of dataset distillation! May I have the opportunity to review the code ? Your assistance is greatly appreciated.
-
Hi, thanks a lot for sharing the training code. If I need to learn new motion patterns with my own datasets which are very different from normal realistic videos, how should I train the model? Do you …
-
Hi @GeorgeCazenavette
I hope all is well. I am wondering if it would be possible for you to upload the Torch tensors containing the distilled dataset for the GLaD paper (CVPR 2023) distillation me…