Enable Left-Censoring for Random Slope Model

gowerc commented 8 months ago

Closes #287

Note that this PR is dependent on #289 being merged first as I needed to use some of the changes that had already been made in that branch (namely enabling jinja rendering of the longitudinal models base file)

Main changes include:

Pulled the csr_matrix_times_vector() code out of the individual model files and into the base longitudinal model file. My thinking is that this simplifies the cognitive load for using this function when writing future longitudinal models
Introduce a new global variable Ypred_log_lik which contains the log likelihood values per observation (this is opposed to log_lik which contains the log-likelihood values per patient
Updated the GSF / SF / RS models to populate the new Ypred_log_lik
Added dedicated unit test to show that left-censoring is working as expected for the RS model. This works by trashing the data for censored values which would break the model if the data wasn't censored. The test shows that the correct model values are still recovered as expected.

github-actions[bot] commented 8 months ago

Code Coverage Summary

Filename                          Stmts    Miss  Cover    Missing
------------------------------  -------  ------  -------  --------------------------------
R/brier_score.R                     166       0  100.00%
R/DataJoint.R                        76       2  97.37%   264, 270
R/DataLongitudinal.R                119       1  99.16%   245
R/DataSubject.R                      69       1  98.55%   124
R/DataSurvival.R                     77       0  100.00%
R/defaults.R                         10       6  40.00%   18-57, 84
R/generics.R                         23       1  95.65%   49
R/JointModel.R                      122       8  93.44%   142-144, 194, 198, 240, 286, 292
R/JointModelSamples.R                55       0  100.00%
R/Link.R                             55       4  92.73%   159-162
R/LinkComponent.R                    47       5  89.36%   100, 118, 132-149
R/LongitudinalGSF.R                  64       0  100.00%
R/LongitudinalModel.R                37      12  67.57%   70-85
R/LongitudinalQuantities.R           85       0  100.00%
R/LongitudinalRandomSlope.R          27       0  100.00%
R/LongitudinalSteinFojo.R            57       8  85.96%   113-135
R/Parameter.R                        14       0  100.00%
R/ParameterList.R                    42       1  97.62%   184
R/Prior.R                           236       8  96.61%   480, 576, 588-606
R/Quantities.R                      105       0  100.00%
R/settings.R                         12      12  0.00%    55-69
R/SimGroup.R                          5       0  100.00%
R/SimJointData.R                     72       1  98.61%   103
R/SimLongitudinal.R                   5       2  60.00%   22, 40
R/SimLongitudinalGSF.R               51       0  100.00%
R/SimLongitudinalRandomSlope.R       42       0  100.00%
R/SimLongitudinalSteinFojo.R         48       0  100.00%
R/SimSurvival.R                     104       0  100.00%
R/StanModel.R                        15       0  100.00%
R/StanModule.R                      179       6  96.65%   199-200, 242, 253, 390, 418
R/SurvivalExponential.R              10       0  100.00%
R/SurvivalLoglogistic.R              11       0  100.00%
R/SurvivalModel.R                    19       0  100.00%
R/SurvivalQuantities.R              155       6  96.13%   178-183
R/SurvivalWeibullPH.R                11       0  100.00%
R/utilities.R                       146       1  99.32%   13
R/zzz.R                              28      25  10.71%   3, 8-12, 14-35, 47
TOTAL                              2399     110  95.41%

Diff against main

Filename      Stmts    Miss  Cover
----------  -------  ------  --------
TOTAL             0       0  +100.00%

Results for commit: 3f21430f018aa5c7559490302ea5871fe003c0ec

Minimum allowed coverage is 80%

:recycle: This comment has been updated with latest results

github-actions[bot] commented 8 months ago

Unit Tests Summary

1 files 40 suites 5m 52s :stopwatch: 122 tests 91 :white_check_mark: 31 :zzz: 0 :x: 853 runs 822 :white_check_mark: 31 :zzz: 0 :x:

Results for commit 3f21430f.

:recycle: This comment has been updated with latest results.

github-actions[bot] commented 8 months ago

Unit Test Performance Difference

Test Suite	$Status$	Time on `main`	$±Time$	$±Tests$	$±Skipped$	$±Failures$	$±Errors$
JointModelSamples	💔	$31.73$	$+1.60$	$0$	$0$	$0$	$0$
LongitudinalQuantiles	💔	$41.16$	$+4.17$	$0$	$0$	$0$	$0$
LongitudinalRandomSlope	💔	$34.32$	$+12.99$	$+3$	$0$	$0$	$0$
SurvivalQuantities	💔	$14.40$	$+2.15$	$0$	$0$	$0$	$0$
SurvivalWeibullPH	💔	$28.60$	$+1.02$	$0$	$0$	$0$	$0$
model_multi_chain	💔	$39.33$	$+3.37$	$0$	$0$	$0$	$0$

Additional test case details

| Test Suite | $Status$ | Time on `main` | $±Time$ | Test Case | |:-----|:----:|:----:|:----:|:-----| | JointModelSamples | 💔 | $31.73$ | $+1.60$ | print_works_as_expected_for_JointModelSamples | | LongitudinalQuantiles | 💔 | $32.91$ | $+2.98$ | Test_that_LongitudinalQuantities_works_as_expected | | LongitudinalRandomSlope | 💔 | $27.10$ | $+2.15$ | LongitudinalRandomSlope_correctly_generates_an_intercept_per_study | | LongitudinalRandomSlope | 👶 | | $+9.88$ | Random_Slope_Model_left_censoring_works_as_expected | | SurvivalQuantities | 💔 | $8.07$ | $+1.13$ | SurvivalQuantities_and_autoplot.SurvivalQuantities_works_as_expected | | SurvivalWeibullPH | 💔 | $28.50$ | $+1.01$ | SurvivalWeibullPH_can_recover_known_values | | model_multi_chain | 💔 | $39.33$ | $+3.37$ | Can_recover_known_distribution_parameters_from_random_slope_model_when_using_multiple_chains |

Results for commit b871162099c4f4cf8ff3cf0d57135c25282280bc

♻️ This comment has been updated with latest results.

Genentech / jmpost