Def data enhancements

[kgoldfeld]

@assignUser This Includes two new functions defRepeat and defRepeatAdd. Tests have been included.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 608c3d7 is merged into main:

•   :ballot_box_with_check:define_data: 25.1ms -> 24.4ms [-7.76%, +2.37%]
•   :ballot_box_with_check:dist_beta: 3.82s -> 3.68s [-9.98%, +2.35%]
•   :ballot_box_with_check:dist_binary: 407ms -> 404ms [-9.69%, +8.44%]
•   :ballot_box_with_check:dist_binomial: 599ms -> 425ms [-92.46%, +34.23%]
•   :ballot_box_with_check:dist_categorical: 883ms -> 904ms [-3.37%, +8.16%]
•   :ballot_box_with_check:dist_exponential: 210ms -> 207ms [-9.26%, +6.31%]
•   :ballot_box_with_check:dist_gamma: 409ms -> 419ms [-5.85%, +10.87%]
•   :ballot_box_with_check:dist_mixture: 4s -> 4.01s [-6.71%, +6.93%]
•   :ballot_box_with_check:dist_normal: 404ms -> 582ms [-49.08%, +137.35%]
•   :ballot_box_with_check:gen_all_dists: 646ms -> 884ms [-52.34%, +126.27%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 87b6811 is merged into main:

•   :ballot_box_with_check:define_data: 31.1ms -> 30.8ms [-4.34%, +2.58%]
•   :ballot_box_with_check:dist_beta: 4.39s -> 4.4s [-1.39%, +1.87%]
•   :ballot_box_with_check:dist_binary: 509ms -> 521ms [-1.94%, +6.4%]
•   :ballot_box_with_check:dist_binomial: 508ms -> 511ms [-2.15%, +3.46%]
•   :ballot_box_with_check:dist_categorical: 1.14s -> 1.12s [-4.17%, +0.65%]
• ❗🐌dist_exponential: 261ms -> 269ms [+0.61%, +5.24%]
•   :ballot_box_with_check:dist_gamma: 517ms -> 528ms [-1.14%, +5.54%]
•   :ballot_box_with_check:dist_mixture: 4.7s -> 4.74s [-0.66%, +2.31%]
•   :ballot_box_with_check:dist_normal: 519ms -> 524ms [-1.78%, +3.7%]
•   :ballot_box_with_check:gen_all_dists: 804ms -> 801ms [-3.08%, +2.28%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[kgoldfeld]

@assignUser This was a tad premature - I haven't added anything to the vignettes - I need to do that.

[kgoldfeld]

@assignUser Vignette is updated, so I think it is ready to go.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if a38e474 is merged into main:

•   :ballot_box_with_check:define_data: 38.1ms -> 37.1ms [-16.26%, +11.1%]
•   :ballot_box_with_check:dist_beta: 4.81s -> 4.8s [-2.67%, +2.45%]
•   :ballot_box_with_check:dist_binary: 547ms -> 542ms [-5.46%, +3.92%]
•   :ballot_box_with_check:dist_binomial: 547ms -> 543ms [-4.08%, +2.64%]
•   :ballot_box_with_check:dist_categorical: 1.2s -> 1.19s [-2.95%, +1.85%]
•   :ballot_box_with_check:dist_exponential: 282ms -> 278ms [-9.21%, +6.95%]
•   :ballot_box_with_check:dist_gamma: 542ms -> 541ms [-2.13%, +1.8%]
•   :ballot_box_with_check:dist_mixture: 5.14s -> 5.06s [-3.9%, +0.9%]
•   :ballot_box_with_check:dist_normal: 538ms -> 545ms [-2.01%, +4.66%]
• ❗🐌gen_all_dists: 827ms -> 843ms [+0.26%, +3.71%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[assignUser]

I'll check it out this weekend :)

[assignUser]

Do we need defRepeatAdd in addition to defRepeat as they do the same thing? Code duplication like this is usally a big no-no :D

[kgoldfeld]

Well - they do the same thing, except for the variable checking, just like defData and defDataAdd. defRepeatAdd calls defDataAdd, whereas defRepeat calls defData.

[assignUser]

Hmm I see. Ah that remindes me of #18
But that can still be done without complete duplication of code... I`ll try it later

[kgoldfeld]

Alternatively, defRepeat wouldn't be needed at all if defData had another argument nVars that would just indicate how many instances of the current variable. But, I don't mind the extra function, since it makes it explicit what the user is trying to do. Again, complete transparency at the expense of efficiency.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if a53e7da is merged into main:

•   :ballot_box_with_check:define_data: 27.5ms -> 27.5ms [-2.36%, +2.11%]
•   :ballot_box_with_check:dist_beta: 4.21s -> 4.25s [-2.59%, +4.54%]
•   :ballot_box_with_check:dist_binary: 456ms -> 459ms [-1.32%, +2.49%]
•   :ballot_box_with_check:dist_binomial: 456ms -> 465ms [-0.95%, +4.85%]
•   :ballot_box_with_check:dist_categorical: 1.04s -> 1.05s [-2.19%, +3.53%]
•   :ballot_box_with_check:dist_exponential: 233ms -> 230ms [-3.7%, +1.09%]
•   :ballot_box_with_check:dist_gamma: 462ms -> 468ms [-0.39%, +3.05%]
•   :ballot_box_with_check:dist_mixture: 4.5s -> 4.53s [-1.58%, +3.15%]
•   :ballot_box_with_check:dist_normal: 461ms -> 465ms [-1.64%, +3.28%]
•   :ballot_box_with_check:gen_all_dists: 718ms -> 710ms [-5.27%, +2.98%]
  Further explanation regarding interpretation and methodology can be found in the documentation.

[kgoldfeld]

@assignUser Just wanted to confirm that we are on board with this, and I can go ahead and merge ...

[assignUser]

Sure, go ahead we will deal with the code duplication when handeling #18 as it is basically the same thing :D