sampling.generate_lhs

generate_lhs(n, ranges, seed=None)

Generates a Latin Hypercube Sample and scales it to the provided variable bounds.

Parameters

Name	Type	Description	Default
n	int	The total number of samples to generate.	required
ranges	Union[pd.DataFrame, Dict]	Definition of input variables. - Dict Format: {‘Name’: (Min, Max), …} e.g. {‘Length’: (0, 10)} - DataFrame Format: Columns [‘Name’, ‘Min’, ‘Max’]	required
seed	int	Sets the random seed for reproducibility. Defaults to None.	`None`

Returns

Name	Type	Description
	pd.DataFrame	pd.DataFrame: A dataframe containing the scaled simulation parameters, where column names correspond to the keys in `ranges`. Returns an empty DataFrame if `ranges` is empty.

Raises

Name	Type	Description
	ValueError	If required columns are missing, types are incorrect, or Min >= Max.
	TypeError	If `ranges` is not a Dictionary or DataFrame.

Examples

Using a dictionary (Recommended):

ranges = {'Length': (0, 10), 'Angle': (0, 90)}
df = generate_lhs(n=3, ranges=ranges, seed=42)
print(df.round(2))
#    Length  Angle
# 0    3.75  85.54
# 1    9.51  13.56
# 2    7.32  54.17

Using a DataFrame (Legacy/Advanced):

import pandas as pd
vars_df = pd.DataFrame([
    {'Name': 'Length', 'Min': 0, 'Max': 10},
    {'Name': 'Angle', 'Min': 0, 'Max': 90}
])
df = generate_lhs(n=3, ranges=vars_df, seed=42)