"""
Generate synthetic two-moons datasets for binary classification tasks.
This module creates multiple configurations of 2D two-moons datasets with
varying sample sizes and noise levels, useful for testing machine learning
algorithms on non-linearly separable data with interleaving classes.
"""
from sklearn.datasets import make_moons
import pandas as pd
import numpy as np
import itertools
import json
import os
# parameters to vary across the configurations
N_SAMPLES = list(range(100, 300, 20))
NOISE = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]
[docs]
def generate_moons_datasets(
n_samples=N_SAMPLES,
noise=NOISE,
save_path=None,
random_state=42,
):
"""
Generate multiple two-moons datasets with varying parameters.
Creates a series of 2D datasets where samples form two interleaving half-circles
(moons), providing a challenging non-linearly separable binary classification problem.
Each configuration varies the number of samples and noise level.
Parameters
----------
n_samples : list of int, default=range(100, 300, 20)
List of sample sizes to generate for each configuration.
noise : list of float, default=[0.1, 0.2, ..., 0.9]
List of noise standard deviations to apply to the data.
save_path : str, optional
Directory path where datasets and configuration files will be saved.
random_state : int, default=42
Random seed for reproducibility.
Returns
-------
None
Saves CSV files for each dataset configuration and a JSON file with
all configuration parameters.
Notes
-----
- Each dataset is saved as 'moons_data-{i}.csv' where i is the configuration number
- Configuration parameters are saved in 'dataset_config.json'
- The last column 'class' contains binary labels (0 or 1)
- Two-moons datasets are commonly used to evaluate algorithms on interleaving patterns
Examples
--------
>>> from qbiocode.data_generation import generate_moons_datasets
>>> generate_moons_datasets(n_samples=[100, 200], noise=[0.1, 0.3], save_path='data')
Generating moons dataset...
"""
print("Generating moons dataset...")
np.random.seed(random_state)
if save_path is None:
save_path = 'moons_data'
if not os.path.exists(save_path):
os.makedirs(save_path)
# enumerate all possible combinations of parameters based on ranges above
configurations = list(itertools.product(*[n_samples, noise]))
# print(configurations)
# print(len(configurations))
count_configs = 1
dataset_config = {}
# populate all the configs with the corresponding argument values
for n_s, n_n in configurations:
config = "n_samples={}, noise={}".format(
n_s, n_n,
)
# print(count_configs)
# iteratively run the function for each combination of arguments
X, y = make_moons(
n_samples=n_s,
noise=n_n,
random_state=random_state,
)
# print("Configuration {}/{}: {}".format(count_configs, len(configurations), config))
dataset = pd.DataFrame(X)
dataset['class'] = y
with open( os.path.join( save_path, 'dataset_config.json' ), 'w') as outfile:
dataset_config.update({'moons_data-{}.csv'.format(count_configs):
{'n_samples': n_s,
'noise': n_n}})
json.dump(dataset_config, outfile, indent=4)
new_dataset = dataset.to_csv( os.path.join( save_path, 'moons_data-{}.csv'.format(count_configs)), index=False)
count_configs += 1
# print(X.shape)
# print(y.shape)
return
