mioXpektron.analysis.compare

Cross-dataset and model-family comparison utilities.

Functions

categorize_model(model_name)

Assign a model name to a coarse family label.

compare_model_results(results_a, results_b, *)

Merge two benchmark tables and compute accuracy deltas.

plot_dataset_model_comparison(comparison_df, ...)

Side-by-side accuracy bars for two datasets.

plot_family_comparison(results_df, savepath, *)

Plot best and mean accuracy per model family.

run_multi_dataset_comparison(datasets, *[, ...])

Run analysis workflows on multiple datasets and compare ML benchmarks.

summarize_model_families(results_df)

Aggregate benchmark metrics by model family.
mioXpektron.analysis.compare.categorize_model(model_name)[source]

Assign a model name to a coarse family label.

Parameters:

model_name (str)

Return type:

str

mioXpektron.analysis.compare.summarize_model_families(results_df)[source]

Aggregate benchmark metrics by model family.

Parameters:

results_df (DataFrame)

Return type:

DataFrame

mioXpektron.analysis.compare.plot_family_comparison(results_df, savepath, *, dataset_name='dataset')[source]

Plot best and mean accuracy per model family.

Parameters:
Return type:

None

mioXpektron.analysis.compare.compare_model_results(results_a, results_b, *, dataset_a='dataset_a', dataset_b='dataset_b')[source]

Merge two benchmark tables and compute accuracy deltas.

Parameters:
Return type:

DataFrame

mioXpektron.analysis.compare.plot_dataset_model_comparison(comparison_df, savepath, *, dataset_a='dataset_a', dataset_b='dataset_b', top_n=15)[source]

Side-by-side accuracy bars for two datasets.

Parameters:
Return type:

None

mioXpektron.analysis.compare.run_multi_dataset_comparison(datasets, *, outdir='comparison_outputs', config=None, run_ml_benchmark=True)[source]

Run analysis workflows on multiple datasets and compare ML benchmarks.

Parameters:
Return type:

Dict[str, Any]