sinapsis-sktime 0.1.0

Module for handling time series data and forecasting using sktime.

Sinapsis SKTime

Module for handling time series data and forecasting using sktime.

🐍 Installation • 🚀 Features • 📚 Usage Example • 📙 Documentation • 🔍 License

Sinapsis SKTime provides a powerful and flexible implementation for time series forecasting using the sktime library.

🐍 Installation

Install using your package manager of choice. We encourage the use of uv

Example with uv:

  uv pip install sinapsis-sktime --extra-index-url https://pypi.sinapsis.tech

or with raw pip:

  pip install sinapsis-sktime --extra-index-url https://pypi.sinapsis.tech

🚀 Features

Templates Supported

The Sinapsis SKTime provides a powerful and flexible implementation for time series forecasting using the sktime library.

ThetaForecaster

The following attributes apply to ThetaForecaster template:

• generic_field_key_data (str, required): The key of the generic field containing the time series data.

This template loads time series data from the specified generic field key and creates a TimeSeriesPacket which is then added to the DataContainer.

SKTime Forecasters

The following attributes apply to all SKTime forecasting templates:

• generic_field_for_data (str, optional): The key of the generic field where datasets are stored. Defaults to "SKTimeDatasets".
• model_save_path (str, required): Path where the trained model will be saved.
• n_steps_ahead (int, optional): Number of steps ahead to forecast. Defaults to 37.

The package provides access to a wide range of forecasting models from sktime, including:

• ARCH
• StatsForecastARCH
• StatsForecastGARCH
• ARIMA
• AutoARIMA
• ExponentialSmoothing
• StatsModelsARIMA
• NaiveForecaster
• NaiveVariance
• ThetaForecaster
• ThetaModularForecaster
• TrendForecaster
• PolynomialTrendForecaster

[!TIP] Use CLI command sinapsis info --all-template-names to show a list with all the available Template names installed with Sinapsis SKTime.

[!TIP] Use CLI command sinapsis info --example-template-config TEMPLATE_NAME to produce an example Agent config for the Template specified in TEMPLATE_NAME.

For example, for ThetaForecaster use sinapsis info --example-template-config ThetaForecasterSKTimeWrapper to produce the following example config:

agent:
  name: my_test_agent
templates:
- template_name: InputTemplate
  class_name: InputTemplate
  attributes: {}
- template_name: ThetaForecasterSKTimeWrapper
  class_name: ThetaForecasterSKTimeWrapper
  template_input: InputTemplate
  attributes:
    root_dir: '`replace_me:<class ''str''>`'
    model_save_path: '`replace_me:<class ''str''>`'
    n_steps_ahead: 37
    thetaforecaster_init:
      initial_level: null
      deseasonalize: true
      sp: 1
      deseasonalize_model: multiplicative

📚 Usage Example

Below is an example configuration for **Sinapsis SKTime** using a Theta forecasting model. This setup loads time series data into a TimeSeriesPacket and applies the Theta model for forecasting.

Example config

agent:
  name: ThetaForecasterAgent
  description: 'Agent for time series forecasting using Theta'

templates:
  - template_name: InputTemplate
    class_name: InputTemplate
    attributes: {}

  - template_name: load_airlineWrapper
    class_name: load_airlineWrapper
    template_input: InputTemplate
    attributes:
      split_dataset: true
      train_size: 0.8
      store_as_time_series: True
      load_airline:
        {}

  - template_name: ThetaForecasterSKTimeWrapper
    class_name: ThetaForecasterSKTimeWrapper
    template_input: load_airlineWrapper
    attributes:
      generic_field_key: load_airlineWrapper
      n_steps_ahead: 12
      forecast_horizon_in_fit: true
      model_save_path: "artifacts/theta_forecaster.pkl"
      task_type: "forecasting"
      thetaforecaster_init:
        sp: 12

This configuration defines an agent and a sequence of templates to handle the data and perform predictions.

To run the config, use the CLI:

sinapsis run name_of_config.yml

📙 Documentation

Documentation for this and other sinapsis packages is available on the sinapsis website

Tutorials for different projects within sinapsis are available at sinapsis tutorials page

🔍 License

This project is licensed under the AGPLv3 license, which encourages open collaboration and sharing. For more details, please refer to the LICENSE file.

For commercial use, please refer to our official Sinapsis website for information on obtaining a commercial license.