From trying to run the MCMC, solution, the LLH cannot be computed. The exact error is given below:
/run/media/jacob/SSD/Development/funfolding/funfolding/solution/likelihood.py:159: UserWarning: Hessian is not tested!
warnings.warn('Hessian is not tested!')
Traceback (most recent call last):
File "/run/media/jacob/SSD/Development/funfolding/examples/02_unfolding_example.py", line 128, in <module>
solution, V_f_est = sol_mini.fit(constrain_N=False)
File "/run/media/jacob/SSD/Development/funfolding/funfolding/solution/_solution.py", line 138, in fit
hess_matrix = self.llh.evaluate_hesse_matrix(solution.x)
File "/run/media/jacob/SSD/Development/funfolding/funfolding/solution/likelihood.py", line 164, in evaluate_hesse_matrix
return ((self.tau * self.C) + H_unreg) * self.factor
TypeError: can't multiply sequence by non-int of type 'NoneType'
The main problem seems to be that self.tau is set to None if nothing is given. This results in the inability to actually calculate the log likelihood.
From trying to run the MCMC, solution, the LLH cannot be computed. The exact error is given below:
The main problem seems to be that self.tau is set to
Noneif nothing is given. This results in the inability to actually calculate the log likelihood.