DATASCIENCEARCHIVE Telegram 124
之前推过一篇 Facebook 的 ConvNeXT,今天又看到一个上个星期新出来的 XFormer,直接把二次的 attention 换成了线性 attention,也是在 ViTs 的基础上想办法减少资源使用。没看到什么别的新奇的地方。主要就是减少 GPU 使用,对数据量要求也低了不少,我没做实验。https://arxiv.org/abs/2201.10271,作者给了代码:https://github.com/pranavphoenix/cxv



tgoop.com/DataScienceArchive/124
Create:
Last Update:

之前推过一篇 Facebook 的 ConvNeXT,今天又看到一个上个星期新出来的 XFormer,直接把二次的 attention 换成了线性 attention,也是在 ViTs 的基础上想办法减少资源使用。没看到什么别的新奇的地方。主要就是减少 GPU 使用,对数据量要求也低了不少,我没做实验。https://arxiv.org/abs/2201.10271,作者给了代码:https://github.com/pranavphoenix/cxv

BY Data Science Archive


Share with your friend now:
tgoop.com/DataScienceArchive/124

View MORE
Open in Telegram


Telegram News

Date: |

You can invite up to 200 people from your contacts to join your channel as the next step. Select the users you want to add and click “Invite.” You can skip this step altogether. Avoid compound hashtags that consist of several words. If you have a hashtag like #marketingnewsinusa, split it into smaller hashtags: “#marketing, #news, #usa. How to Create a Private or Public Channel on Telegram? In handing down the sentence yesterday, deputy judge Peter Hui Shiu-keung of the district court said that even if Ng did not post the messages, he cannot shirk responsibility as the owner and administrator of such a big group for allowing these messages that incite illegal behaviors to exist. A vandalised bank during the 2019 protest. File photo: May James/HKFP.
from us


Telegram Data Science Archive
FROM American