| """Prepare exactly the unique input rows used by the evaluator.""" | |
| import argparse | |
| import json | |
| from pathlib import Path | |
| import pandas as pd | |
| def prepare_inputs(frame): | |
| config = json.loads((Path(__file__).parent / 'task.json').read_text()) | |
| columns = ['id'] + config['input_columns'] | |
| if not set(columns) <= set(frame.columns): | |
| raise ValueError('Missing input columns: ' + ', '.join(columns)) | |
| if frame.id.isna().any() or frame.id.duplicated().any(): | |
| raise ValueError('Input IDs must be present and unique.') | |
| return frame[columns].drop_duplicates(config['input_columns'], keep='first').reset_index(drop=True) | |
| if __name__ == '__main__': | |
| parser = argparse.ArgumentParser(description=__doc__) | |
| parser.add_argument('--input', required=True) | |
| parser.add_argument('--output', required=True) | |
| args = parser.parse_args() | |
| frame = prepare_inputs(pd.read_parquet(args.input)) | |
| frame.to_parquet(args.output, index=False) | |
| print(f'Wrote {len(frame)} unique inputs to {args.output}') | |