xie_x1andClaude Sonnet 5
2683fd3576
MC generation -> pile-up assembly -> CNN vs. classical eta-interpolation (scored against MC ground truth) -> optional training -> qualitative inference on real measurement data, self-contained and runnable from a single conda env (see README.md). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
53 lines
2.1 KiB
Python
53 lines
2.1 KiB
Python
"""
|
|
One-time utility (not part of the 01-05 pipeline): extracts a small,
|
|
illustrative slice of real measurement data into data/measurement_samples/,
|
|
so 05_infer_measurement.py has something to fall back to on a machine
|
|
without access to group storage
|
|
(/mnt/sls_det_storage/moench_data/MLXID/Samples/Measurement). Needs that
|
|
access itself to run. ~30 MB total across both targets and all three
|
|
photon counts -- a few tens of thousands of events, nowhere near enough
|
|
for a real reconstruction, just enough to demonstrate the script runs and
|
|
show roughly the right shape of output.
|
|
|
|
Needs: numpy, h5py only.
|
|
"""
|
|
from pathlib import Path
|
|
import h5py
|
|
|
|
SOURCE_DIR = Path('/mnt/sls_det_storage/moench_data/MLXID/Samples/Measurement')
|
|
OUTPUT_DIR = Path(__file__).resolve().parent.parent / 'data/measurement_samples'
|
|
|
|
TARGETS = {
|
|
'Edge2Filters_12keV': '2603MaxIV_Edge2Filters_12keV',
|
|
'Flat2Filters_12keV': '2603MaxIV_Flat2Filters_12keV',
|
|
}
|
|
### (cluster-file prefix, events to keep) -- sized so 2 targets x 3 photon counts totals ~30 MB
|
|
PHOTON_PATTERNS = [
|
|
('1Photon_CS3', 65_000),
|
|
('2Photon_CS6', 32_000),
|
|
('3Photon_CS9', 18_000),
|
|
]
|
|
|
|
if __name__ == '__main__':
|
|
OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
|
|
for target, folder in TARGETS.items():
|
|
out_target_dir = OUTPUT_DIR / target
|
|
out_target_dir.mkdir(exist_ok=True)
|
|
for prefix, n_events in PHOTON_PATTERNS:
|
|
src = SOURCE_DIR / folder / f'{prefix}_chunk0.h5'
|
|
if not src.exists():
|
|
print(f'SKIP (not found): {src}')
|
|
continue
|
|
with h5py.File(src, 'r') as f:
|
|
total = f['clusters'].shape[0]
|
|
n = min(n_events, total)
|
|
clusters = f['clusters'][:n]
|
|
ref_points = f['referencePoint'][:n]
|
|
|
|
dst = out_target_dir / f'{prefix}_chunk0.h5'
|
|
with h5py.File(dst, 'w') as out:
|
|
out['clusters'] = clusters
|
|
out['referencePoint'] = ref_points
|
|
print(f'{src} ({total:,} total) -> {dst} ({n:,} events, {dst.stat().st_size / 1e6:.1f} MB)')
|
|
print('Done.')
|