
Mean and Std for totalReward:
HyperRNN: Mean = 82.5563, Std = 34.2490
RNNEXP: Mean = 57.0197, Std = 29.9001
RNNIMP: Mean = 61.0899, Std = 31.5144

Tukey HSD results for totalReward:
  Multiple Comparison of Means - Tukey HSD, FWER=0.05   
========================================================
 group1  group2 meandiff p-adj   lower    upper   reject
--------------------------------------------------------
HyperRNN RNNEXP -25.5365    0.0 -28.2716 -22.8014   True
HyperRNN RNNIMP -21.4664    0.0 -24.2015 -18.7313   True
  RNNEXP RNNIMP   4.0701 0.0014    1.335   6.8052   True
--------------------------------------------------------

Pairwise p-values:
[9.19375687e-13 9.19375687e-13 1.42154450e-03]

Mean and Std for totalSteps:
HyperRNN: Mean = 77.7513, Std = 8.8539
RNNEXP: Mean = 79.8233, Std = 6.5271
RNNIMP: Mean = 79.9273, Std = 5.9685

Tukey HSD results for totalSteps:
Multiple Comparison of Means - Tukey HSD, FWER=0.05 
====================================================
 group1  group2 meandiff p-adj  lower  upper  reject
----------------------------------------------------
HyperRNN RNNEXP    2.072   0.0  1.4532 2.6908   True
HyperRNN RNNIMP    2.176   0.0  1.5572 2.7948   True
  RNNEXP RNNIMP    0.104 0.918 -0.5148 0.7228  False
----------------------------------------------------

Pairwise p-values:
[9.34807787e-13 9.20152843e-13 9.17978827e-01]

Mean and Std for totalCollisions:
HyperRNN: Mean = 0.0287, Std = 0.1669
RNNEXP: Mean = 0.0287, Std = 0.1669
RNNIMP: Mean = 0.0327, Std = 0.1778

Tukey HSD results for totalCollisions:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
 group1  group2 meandiff p-adj   lower  upper  reject
-----------------------------------------------------
HyperRNN RNNEXP      0.0    1.0 -0.0146 0.0146  False
HyperRNN RNNIMP    0.004 0.7969 -0.0106 0.0186  False
  RNNEXP RNNIMP    0.004 0.7969 -0.0106 0.0186  False
-----------------------------------------------------

Pairwise p-values:
[1.         0.79690494 0.79690494]

Mean and Std for totalBoundary:
HyperRNN: Mean = 0.0093, Std = 0.0962
RNNEXP: Mean = 0.0113, Std = 0.1059
RNNIMP: Mean = 0.0087, Std = 0.0927

Tukey HSD results for totalBoundary:
 Multiple Comparison of Means - Tukey HSD, FWER=0.05 
=====================================================
 group1  group2 meandiff p-adj   lower  upper  reject
-----------------------------------------------------
HyperRNN RNNEXP    0.002 0.8431 -0.0064 0.0104  False
HyperRNN RNNIMP  -0.0007 0.9812 -0.0091 0.0078  False
  RNNEXP RNNIMP  -0.0027 0.7385 -0.0111 0.0058  False
-----------------------------------------------------

Pairwise p-values:
[0.84314277 0.98120949 0.73848033]
