Implementation of Δ48 standardization about d47crunch HOT 6 CLOSED

NB: for now, the following ugly hack does the job.

from D47crunch import *

rawdata = D47data()
rawdata.Nominal_D48 = {
	'ETH-1': 0.138,
	'ETH-2': 0.138,
	'ETH-3': 0.270,
	'ETH-4': 0.223,
	}

rawdata.read('D48hack.csv')
rawdata.wg()
rawdata.crunch()

# temporarily replace all 47 values with 48 values
rawdata._Nominal_D47 = rawdata.Nominal_D47.copy()
rawdata.Nominal_D47 = rawdata.Nominal_D48.copy()

for r in rawdata:
	r['_D47raw'] = r['D47raw']
	r['_d47'] = r['d47']
	r['D47raw'] = r['D48raw']
	r['d47'] = r['d48']

rawdata.refresh()
rawdata.standardize()

D48_results = {}
D48_results['r_D48'] = rawdata.repeatability['r_D47']
D48_results['r_D48a'] = rawdata.repeatability['r_D47a']
D48_results['r_D48u'] = rawdata.repeatability['r_D47u']
for u in rawdata.unknowns:
	D48_results[u] = {
		'D48': rawdata.unknowns[u]['D47'],
		'SE_D48': rawdata.unknowns[u]['SE_D47'],
		}

print()
for k in D48_results:
	print(f'{k:>8}: {D48_results[k]}')

This should generate the following output:

   r_D48: 0.1237838550727198
  r_D48a: 0.13195328760301325
  r_D48u: 0.10585573189123085
   DVH-2: {'D48': 0.2575489652935696, 'SE_D48': 0.03312408958860156}
   LGB-2: {'D48': 0.30499761247804147, 'SE_D48': 0.03712619158271187}

D48hack.csv

from d47crunch.

Just pushed a new branch, D48, which defines a D48data class analogous to D47data. This is yet another approach, based on the idea that in practice, Δ47 and Δ48 standardizations are mathematically independent (granted, scrambling corrections are likely to be the same, but let's ignore that for now).

The new branch seems to work as intended when doing this:

from D47crunch import *

rawdata = D47data()
rawdata.read('D48test.csv')
rawdata.wg()
rawdata.crunch()
rawdata.standardize()
rawdata.summary()
rawdata.table_of_sessions()
rawdata.table_of_samples()
rawdata.table_of_analyses()
rawdata.plot_sessions()

rawdata = D48data(rawdata)
rawdata.standardize()
rawdata.summary()
rawdata.table_of_sessions()
rawdata.table_of_samples()
rawdata.table_of_analyses()
rawdata.plot_sessions()

Please feel free to experiment and report any issues.

from d47crunch.

Hi!

I have been using the new version of the code and everything seems to run smoothly. However, today I run into a problem switching to the ""indep_sessions" method.
Here the traceback:

"Traceback (most recent call last):
File "/Users/mattiatag/Documents/Python Works/Standards Monitoring/Reference Frame.py", line 236, in
corrected = correction(path)
File "/Users/mattiatag/Documents/Python Works/Standards Monitoring/Reference Frame.py", line 31, in correction
clumpy.standardize(method = "indep_sessions")
File "/Users/mattiatag/.conda/envs/Science/lib/python3.9/site-packages/D4xCrunch/init.py", line 439, in newfun
out = oldfun(*args, **kwargs)
File "/Users/mattiatag/.conda/envs/Science/lib/python3.9/site-packages/D4xCrunch/init.py", line 1161, in standardize
r[f'wD{self._4x}'] /= (a + a2 * r['t'])
KeyError: 'wD47'

Process finished with exit code 1

from d47crunch.

@MattiaTag can you try again with the current version (d4288ca) of the dev branch? I believe this bug was recently corrected.

from d47crunch.

Yes, it works fine in the new version!

from d47crunch.

Closed following v2.0 release.

from d47crunch.