timeseries-cv 0.1.3

Timeseries cross-validation for Neural Networks

Time-Series Cross-Validation

This python package aims to implement Time-Series Cross Validation Techniques.

The idea is given a training dataset, the package will split it into Train, Validation and Test sets, by means of either Forward Chaining, K-Fold or Group K-Fold.

As parameters the user can not only select the number of inputs (n_steps_input) and outputs (n_steps_forecast), but also the number of samples (n_steps_jump) to jump in the data to train.

The best way to install the package is as follows:

pip install git+https://github.com/DidierRLopes/TimeSeriesCrossValidation

---

This can be seen more intuitively using the jupyter notebook: "example.ipynb" Below you can find an example of the usage of each function for the following Time-Series:

timeSeries = array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26])

---

Split Train

split_train

from TimeSeriesCrossValidation.splitTrain import split_train

X, y = split_train(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

split_train_variableInput

from TimeSeriesCrossValidation.splitTrain import split_train_variableInput

X, y = split_train_variableInput(timeSeries, minSamplesTrain=10, n_steps_forecast=3, n_steps_jump=3)

---

Split Train Val

split_train_val_forwardChaining

from TimeSeriesCrossValidation.splitTrainVal import split_train_val_forwardChaining

X, y, Xcv, ycv = split_train_val_forwardChaining(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

split_train_val_kFold

from TimeSeriesCrossValidation.splitTrainVal import split_train_val_kFold

X, y, Xcv, ycv = split_train_val_kFold(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

split_train_val_groupKFold

from TimeSeriesCrossValidation.splitTrainVal import split_train_val_groupKFold

X, y, Xcv, ycv = split_train_val_groupKFold(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

---

Split Train Val Test

split_train_val_test_forwardChaining

from TimeSeriesCrossValidation.splitTrainValTest import split_train_val_test_forwardChaining

X, y, Xcv, ycv, Xtest, ytest = split_train_val_test_forwardChaining(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

split_train_val_test_kFold

from TimeSeriesCrossValidation.splitTrainValTest import split_train_val_test_kFold

X, y, Xcv, ycv, Xtest, ytest = split_train_val_test_kFold(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)

split_train_val_test_groupKFold

from TimeSeriesCrossValidation.splitTrainValTest import split_train_val_test_groupKFold

X, y, Xcv, ycv, Xtest, ytest = split_train_val_test_groupKFold(timeSeries, n_steps_input=4, n_steps_forecast=3, n_steps_jump=2)