update the example in the README #193
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
|
|
||
| ```julia | ||
| using LogDensityProblems | ||
| import LogDensityProblems |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| import LogDensityProblems | |
| using LogDensityProblems: LogDensityProblems |
| ``` | ||
|
|
||
| Since the support of `x` is constrained to be positive and VI is best done in the unconstrained Euclidean space, we need to use a *bijector* to transform `x` into unconstrained Euclidean space. We will use the [`Bijectors.jl`](https://github.com/TuringLang/Bijectors.jl) package for this purpose. | ||
| Since the support of `σ` is constrained to be positive and most VI algorithms assume an unconstrained Euclidean support, we need to use a *bijector* to transform `θ`. |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| Since the support of `σ` is constrained to be positive and most VI algorithms assume an unconstrained Euclidean support, we need to use a *bijector* to transform `θ`. | |
| Since the support of `σ` is constrained to be positive and most VI algorithms assume an unconstrained Euclidean support, we need to use a *bijector* to transform `θ`. |
|
|
||
| ```julia | ||
| using Bijectors | ||
| import Bijectors |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| import Bijectors | |
| using Bijectors: Bijectors |
Comment on lines
+85
to
+86
| import OpenML | ||
| import DataFrames |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| import OpenML | |
| import DataFrames | |
| using OpenML: OpenML | |
| using DataFrames: DataFrames |
| X = Matrix{Float64}(data[:, 1:(end - 1)]) | ||
| y = Vector{Bool}(data[:, end] .== "Mine") | ||
| ``` | ||
| Let's apply some basic pre-processing and add an intercept column: |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| Let's apply some basic pre-processing and add an intercept column: | |
| Let's apply some basic pre-processing and add an intercept column: |
| d = LogDensityProblems.dimension(model_ad) | ||
| q = MeanFieldGaussian(zeros(d), Diagonal(ones(d))) | ||
| ``` | ||
| The bijector can now be applied to `q` to match the support of the target problem. |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| The bijector can now be applied to `q` to match the support of the target problem. | |
| The bijector can now be applied to `q` to match the support of the target problem. |
| q = MeanFieldGaussian(zeros(d), Diagonal(ones(d))) | ||
| ``` | ||
| The bijector can now be applied to `q` to match the support of the target problem. | ||
| ```julia |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| ```julia | |
| ```julia |
|
|
||
| # Run inference | ||
| ``` | ||
| We can now run VI: |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| We can now run VI: | |
| We can now run VI: |
| # Run inference | ||
| ``` | ||
| We can now run VI: | ||
| ```julia |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| ```julia | |
| ```julia |
Comment on lines
136
to
141
| q_avg, info, _ = AdvancedVI.optimize( | ||
| alg, | ||
| max_iter, | ||
| model, | ||
| model_ad, | ||
| q_transformed; | ||
| ) |
Contributor
There was a problem hiding this comment.
[JuliaFormatter] reported by reviewdog 🐶
Suggested change
| q_avg, info, _ = AdvancedVI.optimize( | |
| alg, | |
| max_iter, | |
| model, | |
| model_ad, | |
| q_transformed; | |
| ) | |
| q_avg, info, _ = AdvancedVI.optimize(alg, max_iter, model_ad, q_transformed;) |
Contributor
|
AdvancedVI.jl documentation for PR #193 is available at: |
Contributor
There was a problem hiding this comment.
Benchmark Results
| Benchmark suite | Current: 4711e40 | Previous: 864f5e9 | Ratio |
|---|---|---|---|
normal/RepGradELBO + STL/meanfield/Zygote |
3884113557 ns |
3940463414.5 ns |
0.99 |
normal/RepGradELBO + STL/meanfield/ReverseDiff |
1087580354 ns |
1080457534 ns |
1.01 |
normal/RepGradELBO + STL/meanfield/Mooncake |
1187372637 ns |
1186415910 ns |
1.00 |
normal/RepGradELBO + STL/fullrank/Zygote |
3876169524.5 ns |
3880348067.5 ns |
1.00 |
normal/RepGradELBO + STL/fullrank/ReverseDiff |
1586673546 ns |
1578889137 ns |
1.00 |
normal/RepGradELBO + STL/fullrank/Mooncake |
1230627189 ns |
1245752131 ns |
0.99 |
normal/RepGradELBO/meanfield/Zygote |
2729318274.5 ns |
2745244233 ns |
0.99 |
normal/RepGradELBO/meanfield/ReverseDiff |
753836095 ns |
779172740 ns |
0.97 |
normal/RepGradELBO/meanfield/Mooncake |
1053213407 ns |
1065162127 ns |
0.99 |
normal/RepGradELBO/fullrank/Zygote |
2746187641.5 ns |
2779479552 ns |
0.99 |
normal/RepGradELBO/fullrank/ReverseDiff |
938784644 ns |
922764899.5 ns |
1.02 |
normal/RepGradELBO/fullrank/Mooncake |
1101792362 ns |
1099112093 ns |
1.00 |
normal + bijector/RepGradELBO + STL/meanfield/Zygote |
5608935852 ns |
5520186124 ns |
1.02 |
normal + bijector/RepGradELBO + STL/meanfield/ReverseDiff |
2347525187 ns |
2315766289 ns |
1.01 |
normal + bijector/RepGradELBO + STL/meanfield/Mooncake |
4004306569 ns |
3960343348 ns |
1.01 |
normal + bijector/RepGradELBO + STL/fullrank/Zygote |
5631136713 ns |
5464591378 ns |
1.03 |
normal + bijector/RepGradELBO + STL/fullrank/ReverseDiff |
2965503150.5 ns |
2899203076.5 ns |
1.02 |
normal + bijector/RepGradELBO + STL/fullrank/Mooncake |
4171209341.5 ns |
4077037587 ns |
1.02 |
normal + bijector/RepGradELBO/meanfield/Zygote |
4287497307.5 ns |
4259257598 ns |
1.01 |
normal + bijector/RepGradELBO/meanfield/ReverseDiff |
2016252830 ns |
1969823950 ns |
1.02 |
normal + bijector/RepGradELBO/meanfield/Mooncake |
3859733151.5 ns |
3921065959 ns |
0.98 |
normal + bijector/RepGradELBO/fullrank/Zygote |
4364530532 ns |
4347716606 ns |
1.00 |
normal + bijector/RepGradELBO/fullrank/ReverseDiff |
2290625728 ns |
2217527377 ns |
1.03 |
normal + bijector/RepGradELBO/fullrank/Mooncake |
4015809201.5 ns |
3950691343.5 ns |
1.02 |
This comment was automatically generated by workflow using github-action-benchmark.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.