Speed up

[gravesti]

From our discussion yesterday @audreyyeoCH about performance problems, I reviewed the code.
I found there is significant potential to speed up some of the distribution calculations. I have adapted some old code of mine implementing beta mixtures.
I believe there is still potential for improvement.

[github-actions]

Code Coverage Summary

Filename                 Stmts    Miss  Cover    Missing
---------------------  -------  ------  -------  ----------------------------------
R/betadiff.R                59       0  100.00%
R/boundsPostprob.R          31      31  0.00%    32-71
R/boundsPredprob.R          28      28  0.00%    36-73
R/dbetabinom.R              81       3  96.30%   32, 62, 136
R/oc2.R                    162     162  0.00%    93-326
R/oc3.R                    146     146  0.00%    91-308
R/ocPostprob.R              99       4  95.96%   239-242
R/ocPostprobDist.R         109       0  100.00%
R/ocPredprob.R             182       4  97.80%   285-288
R/ocPredprobDist.R         235       3  98.72%   346-348
R/ocRctPostprobDist.R      154       4  97.40%   236-239
R/ocRctPredprobDist.R      283      55  80.57%   207-222, 225-240, 373-376, 403-421
R/plotBeta.R               103      90  12.62%   60-167
R/plotBounds.R              52      52  0.00%    33-90
R/plotDecision.R            79      79  0.00%    16-136
R/plotOc.R                  15      15  0.00%    16-36
R/postprob.R                34       1  97.06%   106
R/postprobDist.R            77       1  98.70%   204
R/predprob.R                24       0  100.00%
R/predprobDist.R           140       1  99.29%   269
R/runShinyPhase1b.R          4       4  0.00%    8-13
R/sumbetadiff.R             63      63  0.00%    24-101
R/sumTable.R                20      20  0.00%    25-48
TOTAL                     2180     766  64.86%

Diff against main

Filename            Stmts    Miss  Cover
----------------  -------  ------  -------
R/dbetabinom.R        +24       0  +1.56%
R/postprob.R           +5       0  +0.51%
R/postprobDist.R       -1       0  -0.02%
TOTAL                 +28       0  +0.46%

Results for commit: 1e8b717

Minimum allowed coverage is 80%

♻️ This comment has been updated with latest results

[github-actions]

Unit Test Performance Difference

Test Suite	$Status$	Time on main	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
ocPostprob	💚	$52.66$	$-28.96$	$0$	$0$	$0$	$0$
ocPostprobDist	💚	$2.07$	$-1.61$	$0$	$0$	$0$	$0$
ocPredprob	💚	$8.34$	$-4.41$	$0$	$0$	$0$	$0$
ocPredprobDist	💚	$451.25$	$-401.40$	$0$	$0$	$0$	$0$
ocRctPostprobDist	💚	$3.14$	$-2.50$	$0$	$0$	$0$	$0$
ocRctPredprobDist	💚	$172.15$	$-144.95$	$0$	$0$	$0$	$0$
predprobDist	💚	$9.55$	$-8.37$	$0$	$0$	$0$	$0$

Additional test case details

Test Suite	$Status$	Time on main	$±Time$	Test Case
dbetabinom	👶		$+0.01$	dbetaMix_handles_edge_cases
dbetabinom	👶		$+0.01$	pbetaMix_works_for_edge_cases
ocPostprob	💚	$20.57$	$-11.84$	ocPostprob_gives_results_that_are_within_range_to_stats_pbinom
ocPostprob	💚	$11.45$	$-6.74$	the_PrEfficacy_increases_with_increase_Efficacy_looks
ocPostprob	💚	$13.17$	$-7.88$	the_PrFutility_increases_with_increase_futility_looks
ocPostprob	💚	$4.18$	$-2.33$	the_sum_of_Eff_Fut_Gray_zone_probabiliy_is_1
ocPredprob	💚	$7.58$	$-4.06$	the_PrEfficacy_increases_with_increase_Efficacy_looks
ocPredprobDist	💚	$108.18$	$-95.46$	ocPredprobDist_gives_correct_result_and_list_when_relativeDelta_FALSE
ocPredprobDist	💚	$126.84$	$-113.30$	ocPredprobDist_gives_correct_result_and_list_when_relativeDelta_TRUE
ocPredprobDist	💚	$48.60$	$-43.32$	ocPredprobDist_gives_higher_PrEfficacy_with_more_efficacy_looks
ocPredprobDist	💚	$167.60$	$-149.31$	ocPredprobDist_gives_higher_PrFutility_with_more_futility_looks
ocRctPostprobDist	💚	$1.50$	$-1.23$	ocRctPostprobDist_gives_higher_PrFutility_with_decreased_pE
ocRctPredprobDist	💚	$68.69$	$-57.87$	ocRctPredprobDist_gives_higher_PrEfficacy_with_increased_pE
ocRctPredprobDist	💚	$68.72$	$-57.87$	ocRctPredprobDist_gives_higher_PrFutility_with_decreased_pE
ocRctPredprobDist	💚	$34.67$	$-29.20$	the_sum_of_Eff_Fut_Gray_zone_probability_is_1
postprob	👶		$+0.01$	postprob_works_with_vector_x
postprobDist	👶		$+0.01$	postprobDist_works_with_vector_x
predprobDist	💚	$1.46$	$-1.20$	h_predprobdist_gives_correct_list
predprobDist	💚	$2.85$	$-2.54$	h_predprobdist_gives_higher_predictive_probability_when_thetaT_is_lower
predprobDist	💚	$2.92$	$-2.61$	predprobDist_gives_higher_predictive_probability_when_thetaT_is_lower_in_a_single_arm_trial
predprobDist	💚	$1.43$	$-1.27$	predprobDist_gives_the_correct_results_in_a_two_arm_study

Results for commit 4db2bbf

♻️ This comment has been updated with latest results.

[github-actions]

Unit Tests Summary

    1 files     13 suites   1m 48s ⏱️
  115 tests   115 ✅ 0 💤 0 ❌
1 232 runs  1 232 ✅ 0 💤 0 ❌

Results for commit 1e8b717.

♻️ This comment has been updated with latest results.

[danielinteractive]

Great, thanks a lot @gravesti !!
Can you double check that the parts you refactor are already unit test covered? (I think so, just checking)

[gravesti]

Great, thanks a lot @gravesti !! Can you double check that the parts you refactor are already unit test covered? (I think so, just checking)

Yes, will do

[gravesti]

I think this is ready to merge now. @danielinteractive @audreyyeoCH

[danielinteractive]

thanks @gravesti , one more question, I would love to avoid the skipchecks thing if it does not save a lot of time...