* Setting all random seeds to  0 *
Loading model: out_models_control/net_algebraic_noise_rep7_best.pt on cuda
 Loading model that has completed (or started) 50 of 50 epochs
  test episode_type: few_shot_human
  batch size: 25
  max eval length: 10
  number of steps: 196600
  best val loss achieved: 0.0165

BIML specs:
 nparams= 1393801
 nlayers_encoder= 3
 nlayers_decoder= 3
 nhead= 8
 hidden_size= 128
 dim_feedforward= 512
 act_feedforward= gelu
 dropout= 0.1
 

Fitting for the best value of p_lapse use log-like...
  Each value is replicated across 100 random runs/permutations
 p_lapse 0.01 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -594.6461 (SD= 3.3326 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.6007
 p_lapse 0.02 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -560.0711 (SD= 3.047 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5657
 p_lapse 0.03 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -539.0181 (SD= 2.8823 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5445
 p_lapse 0.04 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -524.8689 (SD= 2.7689 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5302
 p_lapse 0.05 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -514.964 (SD= 2.6833 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5202
 p_lapse 0.06 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -507.9546 (SD= 2.6145 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5131
 p_lapse 0.07 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -503.0598 (SD= 2.5571 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5081
 p_lapse 0.08 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -499.7858 (SD= 2.5074 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5048
 p_lapse 0.09 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -497.8005 (SD= 2.4635 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5028
 p_lapse 0.1 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -496.8696 (SD= 2.4238 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.5019
 p_lapse 0.2 :
* Setting all random seeds to  0 *
  run 0
  run 20
  run 40
  run 60
  run 80
  loglike: M = -521.7481 (SD= 2.1375 , Nrep= 100 ) for 990 symbol predictions
    ave LL:  -0.527
* BEST FIT * p_lapse= 0.1 with loglike score of -496.8696
