Source code for scalerqec.Stratified.Scaler

#An updated version of the main method
import numpy as np
import time

from scalerqec.Stratified.stratifiedScurveLER import format_with_uncertainty





[docs] class Scaler: """ Use stratified sampling to estimate the logical error rate of a quantum error correction code. The only hyper parameters are physical error rate and time budget (in seconds). """ def __init__(self, error_rate=0, time_budget=7200): self._error_rate = error_rate self._time_budget = time_budget self._subspace_LE_count={} self._subspace_sample_count={}
[docs] def calculate_LER_from_file(self,filepath,pvalue,codedistance,figname,titlename, repeat=1): pass