Skip to content

Results and Formatting

DataFrame Results

Both rthor.test() and rthor.compare() return pandas DataFrames containing all test results and metadata.

rthor.test() Returns

A single DataFrame with columns:

  • matrix: Matrix identifier (1-indexed)
  • predictions: Number of hypothesized predictions
  • agreements: Number of predictions satisfied
  • ties: Number of tied correlations
  • ci: Correspondence Index (-1 to +1)
  • p_value: Randomization test p-value
  • label: Descriptive label for matrix
  • n_permutations: Number of permutations tested
  • n_variables: Number of variables per matrix

Each row represents one tested matrix. Metadata columns (n_permutations, n_variables) are repeated for tidy data principles.

rthor.compare() Returns

A tuple of two DataFrames: (individual_results, pairwise_comparisons)

individual_results has the same structure as test() output.

pairwise_comparisons DataFrame columns:

  • matrix1, matrix2: Matrix identifiers being compared
  • both_agree: Predictions satisfied by both matrices
  • only1: Predictions satisfied only by matrix 1
  • only2: Predictions satisfied only by matrix 2
  • neither: Predictions satisfied by neither
  • ci: Comparison CI (positive means matrix 2 fits better)
  • p_value: Statistical significance of difference
  • n_permutations, n_variables: Metadata columns

Formatting Functions

Optional functions for compact, interpretable result display:

Print compact, interpreted output of rthor.test() results with CI interpretation and significance:

import rthor

df = rthor.test(data, order="circular6")
rthor.print_results(df)  # Pretty-print the results

Or use the convenience parameter:

df = rthor.test(data, order="circular6", print_results=True)

Print compact, interpreted output of comparison results with winner determination:

individual, pairwise = rthor.compare(data, order="circular6")
rthor.print_comparison(individual, pairwise)

Or use the convenience parameter:

individual, pairwise = rthor.compare(
    data, order="circular6", print_results=True
)

Both functions support rich formatting (if the rich package is installed) or plain text.

Working with DataFrames

Since results are standard pandas DataFrames, you can use all pandas functionality:

# Filter significant results
significant = df[df['p_value'] < 0.05]

# Export to various formats
df.to_csv("results.csv")
df.to_excel("results.xlsx")
df.to_latex("results.tex")

# Get specific values
n_perms = df['n_permutations'].iloc[0]
mean_ci = df['ci'].mean()

# Plotting
import matplotlib.pyplot as plt
df.plot.bar(x='label', y='ci')

rthor.formatting.print_results 

print_results(results: DataFrame, *, use_rich: bool = True) -> None

Print RTHOR test results in a compact, interpretable format.

Source code in src/rthor/formatting.py 
10
11
12
13
14
15
def print_results(results: pd.DataFrame, *, use_rich: bool = True) -> None:
    """Print RTHOR test results in a compact, interpretable format."""
    if use_rich and is_installed("rich"):
        _print_results_rich(results)
    else:
        _print_results_plain(results)

:::

rthor.formatting.print_comparison 

print_comparison(individual: DataFrame, pairwise: DataFrame, *, use_rich: bool = True) -> None

Print comparison results in a compact, interpretable format.

Source code in src/rthor/formatting.py 
158
159
160
161
162
163
164
165
166
167
168
def print_comparison(
    individual: pd.DataFrame,
    pairwise: pd.DataFrame,
    *,
    use_rich: bool = True,
) -> None:
    """Print comparison results in a compact, interpretable format."""
    if use_rich and is_installed("rich"):
        _print_comparison_rich(individual, pairwise)
    else:
        _print_comparison_plain(individual, pairwise)

:::