import time
import numpy as np
from deerlab.utils import der_snr
from deerlab import noiselevel
import logging
import datetime
[docs]
log = logging.getLogger('interface')
class Criteria:
"""
A class for defining criteria for terminating experiments. This should
only be subclassed and not used directly.
"""
def __init__(
self, name: str, test, description: str = '', end_signal=None) -> None:
self.name = name
self.description = description
self.test = test
self.end_signal = end_signal
pass
def __add__(self, __o:object):
if not isinstance(__o,Criteria):
raise RuntimeError("Only objects of the `Criteria` class can be summed together")
new_name = self.name +' + ' + __o.name
new_desc = self.description + ' + ' + __o.description
def new_func(data, verbosity=0):
test1 = self.test(data,verbosity)
test2 = __o.test(data,verbosity)
test = test1 or test2
test_msg = f"Test {new_name}: {test}"
log.debug(test_msg)
return test
if callable(self.end_signal) and callable(__o.end_signal):
def end_signal():
self.end_signal()
__o.end_signal()
elif callable(self.end_signal):
end_signal = self.end_signal
elif callable(__o.end_signal):
end_signal = __o.end_signal
else:
end_signal = None
new_crit = Criteria(new_name,new_func,new_desc,end_signal=end_signal)
return new_crit
class TimeCriteria(Criteria):
def __init__(
self, name: str, end_time: float, description: str = '',
night_hours :tuple = None,**kwargs) -> None:
"""Criteria testing for a specific finishing time. The finishing time
is given as absolute time in the locale of the computer, it is *not*
how the long the measurment continues for.
Parameters
----------
name : str
Name of the criteria
end_time : float
Finishing time in seconds since epoch
description : str, optional
A description of the criteria, by default None
night_hours : tuple, optional
A tuple of two integers specifying the start and end of night hours.
The criteria will always return False during these hours, by default None
"""
log.debug(f"Creating TimeCriteria with end_time: {datetime.datetime.fromtimestamp(end_time).strftime('%Y-%m-%d %H:%M:%S')}")
def test_func(Data, verbosity=0):
now = time.time()
if night_hours is not None:
start, end = night_hours
now_struct = datetime.datetime.fromtimestamp(now)
if now_struct.hour >= start and now_struct.hour < end:
return False
return now > end_time
super().__init__(name, test_func, description,**kwargs)
class SNRCriteria(Criteria):
def __init__(
self, SNR_target: int, description: str = '',verbosity=0,**kwargs) -> None:
"""Criteria testing for signal to noise ratio. This checks the SNR of
the normalised absolute data using the deerlab SNR noise estimation
which is based on the work by Stoher et. al. [1]
Parameters
----------
name : str
_description_
SNR_target : int
The mimimum SNR value.
description : str, optional
_description_, by default None
References
-----------
[1] F. Stoehr, R. White, M. Smith, I. Kamp, R. Thompson, D. Durand,
W. Freudling, D. Fraquelli, J. Haase, R. Hook, T. Kimball, M. Kummel,
K. Levay, M. Lombardi, A. Micol, T. Rogers DERSNR: A Simple & General
Spectroscopic Signal-to-Noise Measurement Algorithm Astronomical Data
Analysis Software and Systems XVII, ASP Conference Series, Vol. 30,
2008, p5.4
"""
def test_func(data, verbosity=verbosity):
# Normalise data
# norm_data = data.data / data.data.max()
# std = der_snr(np.abs(norm_data))
# snr = 1/std
snr=data.epr.correctphase.epr.SNR
test = snr > SNR_target
test_msg = f"Test {self.name}: {test}\t - SNR:{snr}"
log.debug(test_msg)
if verbosity>1:
print(test_msg)
return test
super().__init__("SNR Criteria", test_func, description,**kwargs)