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

Feature list for 0.1.0 release #2

Closed
19 tasks done
storopoli opened this issue Nov 6, 2021 · 3 comments · Fixed by #23
Closed
19 tasks done

Feature list for 0.1.0 release #2

storopoli opened this issue Nov 6, 2021 · 3 comments · Fixed by #23
Assignees
Labels
enhancement New feature or request
Milestone

Comments

@storopoli
Copy link
Member

storopoli commented Nov 6, 2021

  • turing_model function:
    turing_model(formula, data;
          family=Normal,
          priors=DefaultPriors(),
          standardize=false)
  • Close interface with DataFrames.jl and CategoricalArrays.jl using the Tables.jl API:
    • Reading the length(unique(x)) of a group-level intercept/slope
    • Auto dummy variable ($K - 1$) with categorical vectors by reading the levels and using the first level as baseline.
  • Likelihoods:
    • Linear Regression: Gaussian and TDist (IdentityLink)
    • Logistic Regression: Bernoulli
    • Count Data: Poisson and NegativeBinomial (LogLink)
  • Priors:
    • Default Priors and Custom Priors
  • Group-level effects (Hierarchical Models):
    • Non-Centered Parameterization and recompute the parameter back in the return statement
    • Varying-Intercept: (1 | group)
  • Case Studies showcasing:
    • @formula syntax
    • random effects: random-intercept
    • likelihoods
    • CategoricalArrays.jl and DataFrames.jl (Tables.jl) interfaces
    • Use the {rstanarm} datasets (check LICENSE) and the sleep dataset from {brms}
@storopoli storopoli added the enhancement New feature or request label Nov 6, 2021
@storopoli storopoli self-assigned this Nov 6, 2021
@storopoli storopoli added this to the 0.1.0 milestone Nov 6, 2021
@storopoli storopoli mentioned this issue Nov 9, 2021
@phipsgabler
Copy link
Member

One of these models embedded as a @submodel in another Turing model would be a cool show case and interaction test, I think.

@storopoli
Copy link
Member Author

Thanks, do you have good resources on @submodel. I want o learn more about it, never used.

@phipsgabler
Copy link
Member

They are unfortunately not at all documented, but the PR should cover the usage.

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

Successfully merging a pull request may close this issue.

2 participants