Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Char Ensemble Result #44

Closed
jerife opened this issue Mar 9, 2022 · 1 comment
Closed

Char Ensemble Result #44

jerife opened this issue Mar 9, 2022 · 1 comment
Assignees
Labels
enhancement New feature or request

Comments

@jerife
Copy link
Member

jerife commented Mar 9, 2022

Issue #30 에서 언급한 것과 같이 Token끼리 결과를 공유하기 어렵다 판단해 각 모델마다 예측한 char값을 기반으로 average를 구한 결과를 공유하겠습니다.

ground-truth: hi i'm jaewoo

ground-truth prediction1 prediction2 prediction1&2
0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1 0.2, 0.2, 0, 0, 0, 0, 0, 0.7, 0.7, 0.7, 0.7, 0.7, 0.7 0.1, 0.1, 0, 0, 0, 0, 0, 0.6, 0.6, 0.6, 0.3, 0.3, 0.3 0.15, 0.15, 0, 0, 0, 0, 0, 0.65, 0.65, 0.65, 0.5, 0.5, 0.5

결과

Model_A = QA task + Roberta base Tokenizer (LB: 0.832)
Model_B = NER task + Deberta base Tokenizer (LB: 0.861)

Model_A * 1 + Model_B * 1 = LB: 0.852
Model_A * 1 + Model_B * 1.5 = LB: 0.858
Model_A * 1 + Model_B * 2 = LB: 0.859
Model_A * 1 + Model_B * 3 = LB: 0.860

성능이 안좋은 모델과 앙상블하여 더 안좋은 결과가 나온 것일 수도 있다고 생각이 들며,
더 좋은 모델들을 앙상블해서 성능을 확인해야할 것 같다고 판단됩니다.

따라서 이번 회의때 실험하신 모델의 가중치와 코드를 공유하는 법을 한번 더 언급하고자합니다.

@jerife jerife self-assigned this Mar 9, 2022
@jerife jerife added the enhancement New feature or request label Mar 9, 2022
@Kingthegarden
Copy link
Contributor

좋은 아이디어로 실험해 주셔서 감사합니다 ㅎ!

@jerife jerife closed this as completed Mar 10, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants