Jackknife plus ab

[MojiFarmanbar]

Implementing the following methods:

• Jackknife plus after bootstrapping
• Jackknife plus minmax
• Ensemble batch prediction interval

In order to run it, Ditributions and StatsBase packages must be added to ConformalPrediction package.

[codecov-commenter]

Codecov Report

Merging #56 (0c360f4) into main (9ad8bdd) will increase coverage by 0.21%. The diff coverage is 100.00%.

:mega: This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@            Coverage Diff             @@
##             main      #56      +/-   ##
==========================================
+ Coverage   98.60%   98.82%   +0.21%     
==========================================
  Files          12       12              
  Lines         502      596      +94     
==========================================
+ Hits          495      589      +94     
  Misses          7        7              

Impacted Files	Coverage Δ
src/conformal_models/conformal_models.jl	92.30% <ø> (ø)
src/conformal_models/transductive_regression.jl	100.00% <100.00%> (ø)

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[MojiFarmanbar]

hey Pat,

you are right, it is not needed. I removed it.

[pat-alt]

More comments from our meeting:

• [x] Keep the replace argument, but default to true
• [x] Add an argument for bootstrap size $B$, default $n/2$
• [x] Revisit EnbPI (and comment out for now)

[MojiFarmanbar]

hey Pat,

I have adjusted the code as we discussed. it seems in MAPIE, they use the default value of 30 for nsampling.

please check all is ok

[pat-alt]

Ok, awesome, I think this all looks good! Will merge in a few moments and then push a new release to General.

So to summarise, this PR contributes:

• Jackknife plus after bootstrapping
• Jackknife minmax after bootstrapping

As discussed, we'll revisit EnbPI in #57

Thanks again!