https://github.com/pharmcat/nestedterms.jl

Nested term concept package.
https://github.com/pharmcat/nestedterms.jl

Last synced: about 13 hours ago
JSON representation

Nested term concept package.

• Host: GitHub
• URL: https://github.com/pharmcat/nestedterms.jl
• Owner: PharmCat
• License: mit
• Created: 2024-03-24T20:04:39.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-03-24T21:03:20.000Z (8 months ago)
• Last Synced: 2024-03-25T21:28:18.568Z (8 months ago)
• Language: Julia
• Size: 90.8 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# NestedTerms.jl

Nested term concept package.

## Install 

```julia

import Pkg

Pkg.add(url = "https://github.com/PharmCat/NestedTerms.jl.git")

```

## Example

```

using NestedTerms

using StatsModels, CSV, DataFrames, CategoricalArrays

rds = CSV.File(joinpath(dirname(pathof(NestedTerms)), "..", "test", "csv",  "rds12x2.csv")) |> DataFrame # DataFrame with nested factor Subject(Sequence)

transform!(rds, :Sequence => categorical, renamecols=false)

transform!(rds, :Subject => categorical, renamecols=false)

# Nested term exploit function term in @formula macros

formula = @formula(Var ~ (Subject in Sequence))

# Replace FunctionTerm to NestedTerm

formula = NestedTerms.process_nt(formula)

# Make schema and apply_schema 

sch    = schema(formula, rds)

as     = apply_schema(formula, sch) $ upper level factor not added to the formula (but it shoud be there)

# Get model matrix for nested term 

rmf, lmf = modelcols(as, rds)

```

## Unsolvable problems

Can't get real `coefnames` and `levels` until `modelcols` not called. `apply_schema` not able to modify NestedTerm because have no appropriate information.