pede_job.py

Go to the documentation of this file.

"""!
@file pede_job.py
 
Run pede alignment over input bin files generated by tracking accumulation
 
Besides the main purpose of running the pede minimizer, we also do other helpful
tasks that are commonly done with alignment.
1. Apply the results from the pede minimizer to a new iteration of the detector
2. Construct the new iteration of the detector so that another round of tracking
    can be done
3. Merge the histogram files generated from the previous round of tracking for
    later analysis/drawing - this merging is only done if those histogram files
    exist and an output histogrma file is defined in the job JSON.
    - The histogram files are looked for by replacing the input _mille.bin suffix
      on the input files with _gblplots.bin which is the format of the outputs
      done by the tracking example.
"""
import os
from hpsmc.alignment import PEDE, ApplyPedeRes, ConstructDetector
 
job.description = 'alignment parameter minimizatin via pede'
 
if not isinstance(job.input_files, (str, dict)):
    raise ValueError('"input_files" must be a dict list of input files or a path to a text file listing the input files')
 
if isinstance(job.input_files, str):
    listing = job.input_files
    job.input_files = dict()
    with open(listing) as lf:
        for line in lf:
            line = line.strip()
            # skip empty lines or ones starting with #
            if line == '' or line.startswith('#'):
                continue
            # non-empty line, have destination be just the name of the file
            job.input_files[line] = os.path.basename(line)
 
pede = PEDE(inputs=job.input_files.keys() if 'no_copy' in job.params else job.input_files.values())
apply_res = ApplyPedeRes()
construct_det = ConstructDetector()
 
job.add([pede, apply_res, construct_det])
 
gbl_plots = [
    pede_bin_fp.replace('_mille.bin', '_gblplots.root')
    for pede_bin_fp in job.input_files.keys()
    if os.path.isfile(pede_bin_fp.replace('_mille.bin', '_gblplots.root'))
    ]
 
merged_gbl_plots = [
    f for f in job.output_files.keys() if f.endswith('root')
    ]
 
if len(gbl_plots) > 0 and len(merged_gbl_plots) > 0:
    # if we found GBL plots and an output ROOT was provided, we attempt to merge
    from hpsmc import hadd
    hist_merge = hadd(
        inputs=gbl_plots,
        outputs=merged_gbl_plots
        )
    job.add(hist_merge)