Refactor: Replace historical information integration method from conc… #1
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Changed the method of integrating historical information features into the diffusion model.
Previously, historical information features were processed through a GNN, then a style was predicted using a
style_predictor, and the result was concatenated withh_bertbefore being fed into the diffusion model.This has been updated to directly use the historical information features as a context condition in the diffusion model, leveraging cross-attention instead of concatenation.
기존에는 historical information feature를 GNN을 통해 처리한 후,
style_predictor를 사용하여 스타일을 추론하고, 이를h_bert와 concat하여 diffusion model에 입력하는 방식을 사용했습니다.이를 변경하여 historical information feature를 diffusion 모델의 context condition으로 직접 사용하고, concat 대신 cross-attention 방식을 활용하도록 수정했습니다.