Spaces:
Running
Running
Download pkg/src/strokecopilot/data/isles24_layout.py from parsa-hke/stroke-copilot: direct link, hf CLI and curl.
- Browser
- Download file 2.34 kB
-
https://huggingface.co/spaces/parsa-hke/stroke-copilot/resolve/main/pkg/src/strokecopilot/data/isles24_layout.py
- Command line
-
hf download hf://spaces/parsa-hke/stroke-copilot/pkg/src/strokecopilot/data/isles24_layout.py
-
curl -L -o isles24_layout.py https://huggingface.co/spaces/parsa-hke/stroke-copilot/resolve/main/pkg/src/strokecopilot/data/isles24_layout.py
2.34 kB
| """Where each ISLES'24 file lives in the archive, as found in Phase 1 (docs/DATA.md).""" | |
| import os | |
| import subprocess | |
| FILES = { | |
| "ncct": "raw_data/{c}/ses-01/{c}_ses-01_ncct.nii.gz", | |
| "cbf": "derivatives/{c}/ses-01/perfusion-maps/{c}_ses-01_space-ncct_cbf.nii.gz", | |
| "cbv": "derivatives/{c}/ses-01/perfusion-maps/{c}_ses-01_space-ncct_cbv.nii.gz", | |
| "mtt": "derivatives/{c}/ses-01/perfusion-maps/{c}_ses-01_space-ncct_mtt.nii.gz", | |
| "tmax": "derivatives/{c}/ses-01/perfusion-maps/{c}_ses-01_space-ncct_tmax.nii.gz", | |
| "lvo": "derivatives/{c}/ses-01/{c}_ses-01_space-ncct_lvo-msk.nii.gz", | |
| "cow": "derivatives/{c}/ses-01/{c}_ses-01_space-ncct_cow-msk.nii.gz", | |
| "lesion": "derivatives/{c}/ses-02/{c}_ses-02_space-ncct_lesion-msk.nii.gz", | |
| "dwi": "derivatives/{c}/ses-02/{c}_ses-02_space-ncct_dwi.nii.gz", | |
| "baseline": "phenotype/{c}/ses-01/{c}_ses-01_demographic_baseline.xlsx", | |
| "outcome": "phenotype/{c}/ses-02/{c}_ses-02_outcome.xlsx", | |
| } | |
| STRAY_LESION = "derivatives/{c}/ses-02/{c}_ses-02_lesion-msk.nii" | |
| DICTIONARY = "clinical_data-description.xlsx" | |
| # What a working copy keeps. Everything else is decoded but not written, which saves about 85 GiB. | |
| WORKING_MEMBERS = ("train/clinical_data-description.xlsx", "train/phenotype/*", "train/derivatives/*", "train/raw_data/*") | |
| WORKING_EXCLUDE = ( | |
| "*_ctp.nii.gz", "*_cta.nii.gz", "*_adc.nii.gz", | |
| "*_ses-01_cbf.nii.gz", "*_ses-01_cbv.nii.gz", "*_ses-01_mtt.nii.gz", "*_ses-01_tmax.nii.gz", | |
| ) # fmt: skip | |
| def path_of(root, case, name): | |
| return os.path.join(root, FILES[name].format(c=case)) | |
| def find_cases(root): | |
| phenotype = os.path.join(root, "phenotype") | |
| return sorted(d for d in os.listdir(phenotype) if os.path.isdir(os.path.join(phenotype, d))) | |
| def missing_files(root, cases): | |
| """{name: number of cases without that file}, for names with at least one case missing.""" | |
| counts = {name: sum(not os.path.exists(path_of(root, case, name)) for case in cases) for name in FILES} | |
| return {name: n for name, n in counts.items() if n} | |
| def pack(parent, tar_path): | |
| """Tar `train/` under parent, uncompressed: the images are already gzipped.""" | |
| subprocess.run(["tar", "-cf", tar_path, "-C", parent, "train"], check=True) | |
| def unpack(tar_path, parent): | |
| subprocess.run(["tar", "-xf", tar_path, "-C", parent], check=True) | |