Time Series Split Package

A Python package for splitting time series data into training and testing sets, preserving the temporal order.

Installation

You can install this package using pip:

pip install time_series_split

Usage

The split_ts function splits time series data into training and testing sets. It handles both pandas.DataFrame and pandas.Series inputs and ensures that the split maintains the temporal order.

Function: split_ts

import pandas as pd

def split_ts(X, y=None, test_size=0.2):
    """
    Splits time series data into training and testing sets.

    Parameters:
    X (pd.DataFrame or pd.Series): Features of the time series.
    y (pd.Series or pd.DataFrame, optional): Target column corresponding to the features. (default: None)
    test_size (float): Proportion of the dataset to use as test set. (default: 0.2)

    Returns:
    If `y` is provided:
    X_train, X_test, y_train, y_test (np.ndarray or pd.DataFrame, np.ndarray or pd.DataFrame, np.ndarray or pd.Series, np.ndarray or pd.Series): Training and testing sets.

    If `y` is not provided:
    X_train, X_test (np.ndarray or pd.DataFrame, np.ndarray or pd.DataFrame): Training and testing sets.
    """

Parameters

• X: Features of the time series. Can be a pandas.DataFrame or pandas.Series.
• y: (Optional) Target column corresponding to the features. Can be a pandas.Series or pandas.DataFrame.
• test_size: Proportion of the dataset to use as the test set (default is 0.2).

Returns

• If y is provided, returns four objects:

  • X_train: Training features.
  • X_test: Testing features.
  • y_train: Training targets.
  • y_test: Testing targets.

  All returned as numpy.ndarray or pandas.DataFrame/pandas.Series.

• If y is not provided, returns two objects:

  • X_train: Training features.
  • X_test: Testing features.

  Both returned as numpy.ndarray or pandas.DataFrame.

Example

Here's how you can use the split_ts function:

import pandas as pd
from time_series_split.split_ts import split_ts

# Sample DataFrame
data = {'date': [2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009],
        'value': [5, 6, 8, 7, 10, 12, 13, 14, 15, 16]}
df = pd.DataFrame(data)

# Splitting data with target
X_train, X_test, y_train, y_test = split_ts(df['date'], df['value'], test_size=0.3)

# Splitting data
train, test = split_ts(df, test_size=0.3)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

For any questions or support, please contact danttis.