Wavy is a time-series manipulation library designed to simplify the pre-processing steps and reliably avoid the problem of data leakage. Its main structure is built on top of Pandas. Explore the docs 📖

You can install Wavy from pip:

pip install wavyts

import numpy as np
import pandas as pd
import wavy
from wavy import models

# Start with any time-series dataframe:
df = pd.DataFrame({'price': np.random.randn(1000)}, index=range(1000))

# Create panels. Each panel is a frame collection.
x, y = wavy.create_panels(df, lookback=3, horizon=1)

# x and y contain the past and corresponding future data.
# lookback and horizon are the number of timesteps.
print("Lookback:", x.num_timesteps)
print("Horizon:", y.num_timesteps)

# Set train-val-test split. Defaults to 0.7, 0.2 and 0.1, respectively.
wavy.set_training_split(x, y)

# Instantiate a model:
model = models.LinearRegression(x, y)
model.score()

💡 Wavy is:

• A resourceful, high-level package with tools for time-series processing, visualization, and modeling.
• A facilitator for time-series windowing that helps reduce boilerplate code and avoid shape confusion.

❗ Wavy is not:

• An efficient, performance-first framework (yet!).
• Primarily focused on models. Processed data can be easily converted to numpy arrays for further exploration.

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make to wavy are greatly appreciated.

If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! ⭐

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

Distributed under the MIT License. See LICENSE.txt for more information.