agentsclimarketplace

自适应多模态特征融合与正则化

Skill ECNU-ICALK/AutoSkill/SkillBank/ConvSkill/chinese_gpt4_8_GLM4.7/自适应多模态特征融合与正则化

实现一个PyTorch模块,用于动态融合RGB和Event特征,并包含正则化项以防止模型过度偏向某一模态。From its SKILL.md

Install
npx -y skills add ECNU-ICALK/AutoSkill --skill 自适应多模态特征融合与正则化

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

SKILL.md

1.5 KB, 367 tokens by cl100k_base, as published. Nobody here has run it

自适应多模态特征融合与正则化

实现一个PyTorch模块,用于动态融合RGB和Event特征,并包含正则化项以防止模型过度偏向某一模态。

Prompt

你是一个PyTorch专家。你的任务是实现一个可复用的技能,用于带有正则化的自适应融合模块。

  1. 定义AdaptiveFusion类,包含用于RGB和Event特征的可学习权重参数。
  2. 实现前向传播逻辑,计算加权融合后的特征。
  3. 实现一个正则化损失项(例如 (weight_rgb + weight_event - 1)^2),用于鼓励权重和接近1,从而避免模型过度偏向某一模态。
  4. 在前向传播中同时返回融合后的特征和正则化损失。
  5. 解释如何在总训练损失中集成这个正则化损失。

Triggers

  • 如何实现AdaptiveFusion模块
  • 如何在AdaptiveFusion中增加正则化
  • 如何防止模态融合过度偏向
  • 动态权重调整与正则化
  • 多模态特征融合代码实现

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.