This python library aims to offer analysis tools for QCD Data

• Computation of effective mass from 2-point correlation function

def effective_mass(t_correlator,t_initial_guess, t_analysis_type, **analysis_kwargs)

• First order energy by C(t) ~ A cosh((t-Nt/2)*E )

def energy_firstorder( t_correlator, t_meff, t_A0, t_E0, t_fitter, t_fitter_params = dict(),  t_plateau_est_acc = 1e-4, t_plateau_area = 2)

• Jackknife (with and without blocking)

def jackknife(t_data, t_n = 1, t_random_leaveout = False, t_num_ran_indices=None, t_blocked = False, t_num_blocks = None)

• Bootstrap (with and without blocking)

def bootstrap(t_data,t_num_leave_outs, t_blocked = False, t_num_blocks = None)

• Blocking

def blocking(t_data, t_num_blocks = 2) # estimator and variances
def var_per_num_blocks(t_data,t_num_blocks_range = None) # block number analysis

• Fitting
  • Least Square Diagonal Approximation

      class DiagonalLeastSquare(t_model,t_abscissa,t_data=None,t_ordinate=None,t_ordinate_var=None,t_analysis_params=None)
  

    * fit: `Fit_Results = DiagonalLeastSquare.fit()`
    * print: `DiagonalLeastSquare.print_result(*args,**kwargs)`
  

  • Least Square Correltated

      class CorrelatedLeastSquare(t_model,t_abscissa,t_data=None,t_ordinate=None,t_ordinate_cov=None,t_analysis_params=None, t_inv_acc=1e-8)
  

    * fit: `Fit_Results = DiagonalLeastSquare.fit()`
    * print: `DiagonalLeastSquare.print_result(*args,**kwargs)`
  

  • Sampled Least Square Diagonal Approximation

      class Sampled_DiagonalLeastSquare(t_model,t_abscissa,t_data,t_analysis_params)
  

    * fit: `Fit_Results = DiagonalLeastSquare.fit()`
    * print: `DiagonalLeastSquare.print_result(*args,**kwargs)`
  

  • Sampled Least Square Correltated

      class Sampled_CorrelatedLeastSquare(t_model,t_abscissa,t_data,t_analysis_params,t_frozen_cov_flag=False,t_inv_acc=1e-8)
  

    * set `t_frozen_cov_flag = True` to compute the convariance matrix once over full data and do not recompute it on every sample. Might increase
    stability and greatly increases speed.
    * fit: `Fit_Results = DiagonalLeastSquare.fit()`
    * print: `DiagonalLeastSquare.print_result(*args,**kwargs)`
  

• Akaike information criterion

def AIC_chisq(t_dof, t_chisq)
def AICc_chisq(t_dof, t_datasize, *AIC_args, **AIC_kwargs)
def AIC_weights(t_AICs)

• dimension transformations
  • mass: def massMeV(t_mlat, t_a, t_mlat_err = None, t_a_err = None)
    • Note: Either both or no err parameter must be given.

• scipy
• itertools
• warnings
• pathlb
• numpy
• matplotlib.pyplot
• aff USQCDs python aff library (Optional)
  • only required if qcdanalysistools.fileIO.aff is used.
  • needs to be installed beforehand

python setup.py install --user
or
sudo python setup.py install

This software package is accessible under the MIT license.