Dataset Viewer
The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
SNU_invivoNMR — noise-preserving denoising 연구 데이터
MRS(자기공명분광) 및 MNIST denoising 연구의 시뮬레이션 데이터셋과 학습된 모델 체크포인트.
코드는 GitHub 에 있습니다: https://github.com/sarahlee-59/SNU_invivoNMR (private)
사용법
이 저장소를 코드 저장소 위에 그대로 풀면 원래 디렉토리 구조가 복원됩니다.
git clone https://github.com/sarahlee-59/SNU_invivoNMR.git
cd SNU_invivoNMR
python -c "
from huggingface_hub import snapshot_download
snapshot_download('seoyeonlee/SNU_invivoNMR', repo_type='dataset', local_dir='.')
"
경로가 코드에서 참조하는 상대 경로(<실험폴더>/data/train.npz 등)와 그대로 일치합니다.
구성
| 종류 | 개수 | 내용 |
|---|---|---|
*.npz |
46 | 시뮬레이션 데이터셋 (train/val, 스펙트럼 + 파라미터) |
*.pt |
19 | 학습된 모델 체크포인트 |
*.npy |
5 | 공분산 행렬 cov_eps.npy, denoised 스펙트럼 등 |
all_parameters.csv |
8 | 시뮬레이션 파라미터 덤프 |
*.mat |
3 | MRS 스펙트럼 원본 |
| MNIST raw | 8 | MNIST 원본 (idx-ubyte, gz) |
총 97 파일, 약 2.7 GiB.
주요 실험 디렉토리
260714_mirage_exp~260722_mirage_new_exp— "Miracle or mirage" 재현 (fid0/fixedref × kernel2/3)260730_mirage_CRLB,260730_mirage_quant— CRLB 및 Fig 3/4 정량화260803_noise_preserving_denoising— b/eps 분해,data/cov_eps.npy(2048² 공분산)260810_mnist,260810_n2n_nppc_mrs— Noise2Noise + NPPCNMR_experiment/MRS_sim_prelim— 50k 시뮬레이션 데이터
라이선스
연구용 데이터입니다. 비상업적 용도로만 사용해 주세요.
- Downloads last month
- 18