def findNLLParams(self, dataFileName):
function = Function()
NLLExp = NLL(function.expPDF, dataFileName)
optimiser = Optimise()
#finds minimum tau for our NLL
paramsGuess = [2.]
paramsJump = [0.01]
paramsAccuracy = [0.00001]
params = optimiser.min(NLLExp.evalNLL, paramsGuess, paramsJump,
paramsAccuracy, [])
tauMin = params[0]
minNll = NLLExp.evalNLL(params)
#finding err of tau in params by finding where the chiSquare
#increases by 0.5 either side of the minimum
tauRootGuess = [tauMin]
tauRootJump = [0.001]
tauRootAccuracy = [0.00001]
tauErrPos = abs(
optimiser.equalTo((minNll + 0.5), NLLExp.evalNLL, tauRootGuess,
tauRootJump, tauRootAccuracy, [])[0] - tauMin)
tauErrNeg = abs(
optimiser.equalTo((minNll +
0.5), NLLExp.evalNLL, [tauMin - tauErrPos],
tauRootJump, tauRootAccuracy, [])[0] - tauMin)
tauErrMean = (tauErrPos + tauErrNeg) / 2.
print("")
print("tau:\t" + str(np.round(tauMin, 6)) + "\n")
print("tauErr:\t+" + str(np.round(tauErrPos, 6)) + "\t-" +
str(np.round(tauErrNeg, 6)))
print("mean:\t" + str(np.round(tauErrMean, 6)))
print("")
#self.plotNLLAroundMin(NLLExp.evalNLL, params, [tauErrPos], [tauErrNeg])
plotter = Plot()
plotter.plotNLLAroundMin(NLLExp.evalNLL, params, [tauErrPos],
[tauErrNeg])
