SamplePDFBase

class pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase

Bases: pybind11_object

Methods Summary

fill_1d_hist(self)

Do the initial filling of the sample for 1d histogram.

fill_2d_hist(self)

Do the initial filling of the sample for 2d histogram.

get_bin_LLH(self, data, mc, w2)

Get the LLH for a bin by comparing the data and MC. The result depends on having previously set the test statistic using pyMaCh3.sample_pdf.SamplePDFFDBase.set_test_stat()

get_likelihood(self)

Get the sample likelihood at the current point in your model space.

reweight(self)

reweight the MC events in this sample.

set_test_stat(self, test_stat)

Set the test statistic that should be used when calculating likelihoods.

Methods Documentation

fill_1d_hist(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase) None

Do the initial filling of the sample for 1d histogram. You will need to override this.

fill_2d_hist(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase) None

Do the initial filling of the sample for 2d histogram. You will need to override this.

get_bin_LLH(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase, data: float, mc: float, w2: float) float
Get the LLH for a bin by comparing the data and MC. The result depends on having previously set the test statistic using pyMaCh3.sample_pdf.SamplePDFFDBase.set_test_stat()
param data:

The data content of the bin.

param mc:

The mc content of the bin

param w2:

The Sum(w_{i}^2) (sum of weights squared) in the bin, which is sigma^2_{MC stats}

get_likelihood(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase) float

Get the sample likelihood at the current point in your model space. You will need to override this.

reweight(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase) None

reweight the MC events in this sample. You will need to override this.

set_test_stat(self: pyMaCh3._pyMaCh3.sample_pdf.SamplePDFBase, test_stat: pyMaCh3._pyMaCh3.sample_pdf.TestStatistic) None
Set the test statistic that should be used when calculating likelihoods.
param test_stat:

The new test statistic to use