https://github.com/didierrlopes/timeseries-cv
Time-Series Cross-Validation Module
https://github.com/didierrlopes/timeseries-cv
cross-validation python timeseries-analysis
Last synced: 10 months ago
JSON representation
Time-Series Cross-Validation Module
- Host: GitHub
- URL: https://github.com/didierrlopes/timeseries-cv
- Owner: DidierRLopes
- License: mit
- Created: 2020-01-26T18:31:29.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2021-12-12T14:30:17.000Z (about 4 years ago)
- Last Synced: 2025-03-18T08:04:58.940Z (10 months ago)
- Topics: cross-validation, python, timeseries-analysis
- Language: Jupyter Notebook
- Homepage:
- Size: 454 KB
- Stars: 44
- Watchers: 3
- Forks: 10
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Funding: .github/FUNDING.yml
- License: LICENSE
Awesome Lists containing this project
README
# 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 timeseries-cv` and then use it with `import tsxv`.
## Features
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 tsxv.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 tsxv.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 tsxv.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 tsxv.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 tsxv.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 tsxv.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 tsxv.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 tsxv.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)
```
## Citation
This module was developed with co-autorship with Filipe Roberto Ramos (https://ciencia.iscte-iul.pt/authors/filipe-roberto-de-jesus-ramos/cv) for his phD thesis entitled "Data Science in the Modeling and Forecasting of Financial timeseries: from Classic methodologies to Deep Learning". Submitted in 2021 to Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal.
APA
`Ramos, F. (2021). Data Science na Modelação e Previsão de Séries Económico-financeiras: das Metodologias Clássicas ao Deep Learning. (PhD Thesis submitted, Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal).`
```
@phdthesis{FRRamos2021,
AUTHOR = {Filipe R. Ramos},
TITLE = {Data Science na Modelação e Previsão de Séries Económico-financeiras: das Metodologias Clássicas ao Deep Learning},
PUBLISHER = {PhD Thesis submitted, Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal},
YEAR = {2021}
}
```