Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Expanding the work on augmentations in a tutorial #81

Open
wants to merge 8 commits into
base: master
Choose a base branch
from

Conversation

theogf
Copy link
Member

@theogf theogf commented Dec 1, 2021

This introduces a new tutorial similar to the classification one using augmentation as in AGP.jl.
It shows both the solution with and without hyperparameters optimization.

I still need to give a brief explanation on how the augmentation works but things are working well otherwise

examples/d-augmented-gp/script.jl Outdated Show resolved Hide resolved
examples/d-augmented-gp/script.jl Outdated Show resolved Hide resolved
examples/d-augmented-gp/script.jl Outdated Show resolved Hide resolved
examples/d-augmented-gp/script.jl Outdated Show resolved Hide resolved
@theogf theogf changed the title [WIP] Expanding the work of augmentation in a tutorial [WIP] Expanding the work on augmentations in a tutorial Dec 1, 2021
theogf and others added 2 commits December 1, 2021 18:17
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
@theogf
Copy link
Member Author

theogf commented Dec 1, 2021

Oh it looks like the new manifest format is not appreciated?

@@ -75,7 +75,7 @@ function expected_loglik(
# size(fs): (length(y), n_points)
fs = √2 * std.(q_f) .* xs' .+ mean.(q_f)
lls = loglikelihood.(lik.(fs), y)
return sum((1 / √π) * lls * ws)
return (1 / √π) * mapreduce(dot, +, eachrow(lls), ws)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
return (1 / π) * mapreduce(dot, +, eachrow(lls), ws)
return mapreduce(dot, +, eachrow(lls), ws) / π

But also, at this point it's getting really hard to glance at the code and understand what's going on.... what do you think of adding a comment to explaining that?

@st--
Copy link
Member

st-- commented Dec 1, 2021

How about sticking to julia 1.6 for the docs in this PR, and we can sort out the upgrade separately ?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants