nd-ball / py-irt

Bayesian IRT models in Python
MIT License
115 stars 41 forks source link

Add model registry, refactor constructors #22

Closed EntilZha closed 2 years ago

EntilZha commented 2 years ago

More details in commit messages, working towards the multidim implementation (which is currently broken...)

codecov[bot] commented 2 years ago

Codecov Report

Merging #22 (3c02418) into master (2b4b86a) will increase coverage by 21.08%. The diff coverage is 64.19%.

Impacted file tree graph

@@             Coverage Diff             @@
##           master      #22       +/-   ##
===========================================
+ Coverage   40.88%   61.96%   +21.08%     
===========================================
  Files           7       12        +5     
  Lines         340      744      +404     
===========================================
+ Hits          139      461      +322     
- Misses        201      283       +82     
Impacted Files Coverage Δ
py_irt/cli.py 0.00% <0.00%> (ø)
py_irt/scoring.py 0.00% <0.00%> (ø)
py_irt/initializers.py 57.50% <57.50%> (ø)
py_irt/models/one_param_logistic.py 56.03% <61.53%> (+30.32%) :arrow_up:
py_irt/models/two_param_logistic.py 62.29% <66.32%> (+40.16%) :arrow_up:
py_irt/io.py 66.66% <66.66%> (ø)
py_irt/models/four_param_logistic.py 80.68% <80.68%> (ø)
py_irt/dataset.py 83.33% <83.33%> (ø)
py_irt/models/abstract_model.py 86.04% <86.04%> (ø)
py_irt/training.py 96.19% <96.19%> (ø)
... and 9 more

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 609a342...3c02418. Read the comment docs.