janweinreich
commented
2 years ago that bug was because I forgot to adjust the number of heavy atom range not being consistent with the molecule I used as input
Closed janweinreich closed 2 years ago
janweinreich
commented
2 years ago that bug was because I forgot to adjust the number of heavy atom range not being consistent with the molecule I used as input
running
MC_all_moves_QM9_with_bias.pyafter two iterationsm the following issue occurs:SMILE: CCCCC1C2=CC(O)=C3C(=C2)OC(C)(C)C2CCC(C)=CC312 Prediction: -0.07053541066125035 SMILE: CCCCCC1=CC(O)=C2C(=C1)OC(C)(C)C1CCC(C)=CC21 Prediction: -0.07049342675600201 Traceback (most recent call last): File "MC_all_moves_QM9_with_bias.py", line 76, in <module> rw.global_random_change(**global_change_params) File "/home/jan/projects/MOLOPT/molopt/bmapqml/chemxpl/random_walk.py", line 543, in global_random_change self.MC_step_all(**other_kwargs) File "/home/jan/projects/MOLOPT/molopt/bmapqml/chemxpl/random_walk.py", line 523, in MC_step_all output.append(self.MC_step(**mc_step_kwargs, replica_id=replica_id)) File "/home/jan/projects/MOLOPT/molopt/bmapqml/chemxpl/random_walk.py", line 481, in MC_step new_tp, prob_balance=randomized_change(changed_tp, change_prob_dict=self.change_list, **self.used_randomized_change_params) File "/home/jan/projects/MOLOPT/molopt/bmapqml/chemxpl/random_walk.py", line 102, in randomized_change inverse_possibilities, total_inverse_prob=random_choice_from_dict(new_tp.possibilities(), change_prob_dict, get_probability_of=inv_proc) File "/home/jan/projects/MOLOPT/molopt/bmapqml/chemxpl/random_walk.py", line 76, in random_choice_from_dict return possibilities[get_probability_of], np.log(corr_prob_choice[get_probability_of]/prob_sum) KeyError: <function add_heavy_atom_chain at 0x7ff8aa3361f0>