Particionamiento con RepeatedKFold

• Ultima modificación: 2023-02-27 | `YouTube <>`__

#
# repite el KFold n_repeats veces usando diferentes aleatorios
#
from sklearn.model_selection import RepeatedKFold

repeatedKFold = RepeatedKFold(
    # --------------------------------------------------------------------------
    # Número de grupos
    n_splits=5,
    # --------------------------------------------------------------------------
    # Número de repeticiones
    n_repeats=2,
    # --------------------------------------------------------------------------
    # Semilla del generador de aleatorios
    random_state=123,
)

repeatedKFold

RepeatedKFold(n_repeats=2, n_splits=5, random_state=123)

from mymodule import plot_schema

y_classes = [0] * 10 + [1] * 10

plot_schema(repeatedKFold, y_classes)

repeatedKFold.get_n_splits()

10

import numpy as np

X = np.array([[1, 2], [3, 4], [5, 6], [7, 8], [9, 10], [11, 12]])
y = np.array([1, 2, 3, 4, 5, 6])

repeatedKFold = RepeatedKFold(n_splits=3, n_repeats=2)

for i, (train_index, test_index) in enumerate(repeatedKFold.split(X)):
    print(f"Fold {i}:")
    print(f"  Train: index={train_index}")
    print(f"  Test:  index={test_index}")
    print()

Fold 0:
  Train: index=[0 2 3 5]
  Test:  index=[1 4]

Fold 1:
  Train: index=[1 3 4 5]
  Test:  index=[0 2]

Fold 2:
  Train: index=[0 1 2 4]
  Test:  index=[3 5]

Fold 3:
  Train: index=[0 1 3 5]
  Test:  index=[2 4]

Fold 4:
  Train: index=[0 2 4 5]
  Test:  index=[1 3]

Fold 5:
  Train: index=[1 2 3 4]
  Test:  index=[0 5]