Datasets:
File size: 1,774 Bytes
152a1c3 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 | #!/usr/bin/env python3
"""Load the task files shipped in this release."""
from __future__ import annotations
import json
from pathlib import Path
RELEASE = Path(__file__).resolve().parents[1]
DATA = RELEASE / "data"
PROPERTIES = (
"dipole",
"gap",
"infrared",
"raman",
"ultraviolet_visible",
"proton_nmr_peaks",
"proton_nmr_integration",
"proton_nmr_atom",
"carbon_nmr_peaks",
"carbon_nmr_degeneracy",
"carbon_nmr_atom",
"electronic",
"magnetic_resonance",
"vibrational",
"all_properties",
)
CONDITIONS = ("correct", "shuffled", "counterfactual")
FIELDS = ("record_id", "masked_smiles", "mask", "properties", "candidate_completed_smiles")
def task_path(property_name: str, condition: str) -> Path:
if property_name not in PROPERTIES:
raise ValueError(f"unknown property: {property_name}")
if condition not in CONDITIONS:
raise ValueError(f"unknown condition: {condition}")
return DATA / property_name / f"{condition}.jsonl"
def load_tasks(property_name: str, condition: str) -> list[dict]:
path = task_path(property_name, condition)
records = [json.loads(line) for line in path.read_text(encoding="utf-8").splitlines() if line.strip()]
for record in records:
missing = [field for field in FIELDS if field not in record]
if missing:
raise ValueError(f"{path.name} record {record.get('record_id')} missing {missing}")
return records
def main() -> None:
counts = {}
for property_name in PROPERTIES:
counts[property_name] = {
condition: len(load_tasks(property_name, condition))
for condition in CONDITIONS
}
print(json.dumps(counts, indent=2))
if __name__ == "__main__":
main()
|