v1.7.0

[jhollway]

Description

Making

• Improved vector printing to be more succinct and suggest print_all()
• Improved printing methods for changing mnet objects

Modifying

• Added delete_changes() for deleting all changes from a changing network
• Improved to_waves() to work with networks that are changing, longitudinal, or both
• Fixed as_diffusion() to not trim off final wave in the report
• Fixed as_diffusion() to return correct E and R compartments for waning diffusion models

Marking

• Improved is_longitudinal() to only check for tie waves and not nodal changes

Measuring

• Improved net_diversity() and node_diversity()
  • No longer offers a cluster option, which can be obtained using over_membership()
  • Added more methods for calculating diversity, including Teachman's, coefficient of variation, and the Gini coefficient
  • Improved net_diversity() and node_diversity() to use and declare methods appropriate for the vector of attributes
• Added net_homophily() and node_homophily() for measuring homophily according to different methods including
  • Krackhardt's EI index of heterophily as well as its inverse as a measure of homophily
  • Yule's Q as a further measure of homophily
  • Geary's C as a measure of homophily for numeric attributes
• Updated documentation for richness, diversity, and homophily measures
• Fixed node_is_latent(), node_is_infected(), node_is_recoverd() to work with changing networks
• Improved node_is_exposed() to work with changing networks and now accepts a time argument

Tutorials

• Fixed description of to_directed() and to_redirected() in the data tutorial

Checklist:

• Documentation
  • DESCRIPTION file version is bumped by the appropriate increment (major, minor, patch)
  • Date in DESCRIPTION is correct
  • Longer functions are commented inline or broken down into helper functions to help debugging
• PR form
  • Title indicates expected version number
  • PR description above and the NEWS.md file are aligned
  • Description above itemizes changes under subsection titles, e.g. "## Data""
  • Closed, fixed, or related issues are referenced and explained in the description above, e.g. "Fixed #0 by adding A"

[jhollway]

@RWKrause , re #115 I've made a first run at a broader net_homophily() function that incorporates Krackhardt's EI/IE index, Yule's Q, Geary's C, etc. Still to do include:

• broader testing/coverage
• missing data treatment
• ensure defaults/reporting are sensible
• any additional measures to implement while we're at it? maybe incorporate degree assortativity (net_assortativity()) and/or Moran's I (net_spatial())?
• update the diversity tutorial in {migraph} to make use of these new functions/functionalities

[RWKrause]

Hey
AWESOME ♥ I would add assortativity, but not by degree but as a function that requires an attribute, so that you can give it the degree but also a numeric attribute.
Cheers
Robert

[jhollway]

Ok, @RWKrause , do you want to propose a generalised assortativity function, and/or implement a clever way to manage missing data across these and other functions? I'll implement them for two-mode networks, and update the tutorials?

[codecov]

Codecov Report

❌ Patch coverage is 32.08333% with 163 lines in your changes missing coverage. Please review.
✅ Project coverage is 56.74%. Comparing base (faddeb7) to head (44508be).
⚠️ Report is 34 commits behind head on main.

Files with missing lines	Patch %	Lines
R/measure_heterogeneity.R	16.19%	88 Missing ⚠️
R/manip_split.R	20.31%	51 Missing ⚠️
R/mark_nodes.R	64.58%	17 Missing ⚠️
R/class_networks.R	0.00%	3 Missing ⚠️
R/manip_as.R	84.61%	2 Missing ⚠️
R/class_models.R	0.00%	1 Missing ⚠️
R/manip_nodes.R	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #122      +/-   ##
==========================================
- Coverage   58.80%   56.74%   -2.06%     
==========================================
  Files          45       45              
  Lines        5734     5912     +178     
==========================================
- Hits         3372     3355      -17     
- Misses       2362     2557     +195     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.