FFTWrapper
Python template · sinapsis_darts_forecasting.templates.time_series_model
Overview
Base Template for Darts Time Series forecasting models.
FFT — inherited documentation
The base class for "local" forecasting models, handling only single univariate time series.
Local Forecasting Models (LFM) are models that can be trained on a single univariate target series only. In Darts, most models in this category tend to be simpler statistical models (such as ETS or FFT). LFMs usually train on the entire target series supplied when calling :func:`fit()` at once. They can also predict in one go with :func:`predict()` for any number of predictions `n` after the end of the training series.
All implementations must implement the `fit()` and `predict()` methods.Fast Fourier Transform Model
This model performs forecasting on a TimeSeries instance using FFT, subsequent frequency filtering (controlled by the `nr_freqs_to_keep` argument) and inverse FFT, combined with the option to detrend the data (controlled by the `trend` argument) and to crop the training sequence to full seasonal periods Note that if the training series contains any NaNs (missing values), these will be filled using :func:`darts.utils.missing_values.fill_missing_values()`.
Parameters ---------- nr_freqs_to_keep The total number of frequencies that will be used for forecasting. required_matches The attributes of pd.Timestamp that will be used to create a training sequence that is cropped at the beginning such that the first timestamp of the training sequence and the first prediction point have matching phases. If the series has a yearly seasonality, include `month`, if it has a monthly seasonality, include `day`, etc. If not set, or explicitly set to None, the model tries to find the pd.Timestamp attributes that are relevant for the seasonality automatically. trend If set, indicates what kind of detrending will be applied before performing DFT. Possible values: 'poly', 'exp' or None, for polynomial trend, exponential trend or no trend, respectively. trend_poly_degree The degree of the polynomial that will be used for detrending, if `trend='poly'`.
Examples -------- Automatically detect the seasonal periods, uses the 10 most significant frequencies for forecasting and expect no global trend to be present in the data:
>>> FFT(nr_freqs_to_keep=10)
Assume the provided TimeSeries instances will have a monthly seasonality and an exponential global trend, and do not perform any frequency filtering:
>>> FFT(required_matches={'month'}, trend='exp')
Simple usage example, using one of the dataset available in darts
>>> from darts.datasets import AirPassengersDataset >>> from darts.models import FFT >>> series = AirPassengersDataset().load() >>> # increase the number of frequency and use a polynomial trend of degree 2 >>> model = FFT( >>> nr_freqs_to_keep=20, >>> trend= "poly", >>> trend_poly_degree=2 >>> ) >>> model.fit(series) >>> pred = model.predict(6) >>> print(pred.values()) [[471.79323146] [494.6381425 ] [504.5659999 ] [515.82463265] [520.59404623] [547.26720705]]
.. note:: `FFT example notebook <https://unit8co.github.io/darts/examples/03-FFT-examples.html>`__ presents techniques that can be used to improve the forecasts quality compared to this simple usage example.
Capabilities
Capabilities: not setInstallation
Install the published package using pip or uv.
pip install sinapsis-darts-forecasting==1.0.0uv pip install sinapsis-darts-forecasting==1.0.0uv add sinapsis-darts-forecasting==1.0.0Configuration
Effective fields from the inspector schema, including nested Pydantic model fields.
Required
Default: null
Optional
Default: null
Optional
Default: 10
Optional
Default: null
Optional
Default: 3
Optional
Default: null
Optional
Default: false
Optional
Default: 10
Configuration example
Generated from documented defaults. Replace any placeholders and complete required nested settings before running.
templates:
- template_name: "FFTWrapper"
class_name: "FFTWrapper"
attributes:
fft_init:
trend: null
nr_freqs_to_keep: 10
required_matches: null
trend_poly_degree: 3
validation_mode: false
forecast_horizon: 10
View original docstring example
The source example references XGBModelWrapper, not FFTWrapper.
agent:
name: my_test_agent
templates:
- template_name: InputTemplate
class_name: InputTemplate
attributes: {}
- template_name: XGBModelWrapper
class_name: XGBModelWrapper
template_input: InputTemplate
attributes:
forecast_horizon: 100
xgbmodel_init:
lags: 30
lags_past_covariates: 30
output_chunk_length: 100
random_state: 42
n_estimators: 200
learning_rate: 0.1
max_depth: 6
For a full list of options use the sinapsis cli: sinapsis info --all-template-names.
If you want to see all available models, please visit:
https://unit8co.github.io/darts/generated_api/darts.models.forecasting.htmlInheritance
sinapsis_darts_forecasting.templates.time_series_model.TimeSeriesModelEnvironment variables
Environment definitions belong to their canonical Python modules.