seasonality

Computes similarity of the words in the input column with the context words.

Usage

Allows to extract seasonality for different date attributes from training data basing on column to pedict

This calculator can be used with the following method:

seasonality

Examples:

  • Extract week/month seasanalities coefficients from sales history

Main Parameters

The bold options represent the default values when the parameters are optional.

  • input_columns list of columns used as input of the calculators

  • output_columns list of columns added by the calculators

  • global (true, false) Should this calculator be performed before data splitting during training for cross-validation

  • steps [optionnal] (training, prediction, postprocessing) List of steps in a pipeline where columns from this calculator are added to the data. Note that when the training option is listed, the calculator is actually added during preprocessing.

  • store_in_model [optionnal] (true, false) Please indicate whether the "calculated" columns by the calculator should be stored in the model or not to avoid recalculating them during prediction. This is only relevant if the calculated columns are added to both training and prediction. Without this parameter, the values will not be stored in the model. The following parameters only make sense if this parameter is set to true.

  • stored_columns [required if store_in_model is true] List indicating the columns to be stored among the output_columns.

  • stored_keys [required if store_in_model is true] List indicating the columns to use for identifying the correct values to join on the data for prediction among the stored values (logically, they are to be chosen from the input_columns).

Specific Parameters

  • value Name of the column to use for extracting seasonality

  • time_col Name of the column that contains time

Examples

  1. We want to capture additional weekly effects on sales. The month column is calculated from the date_attribute. Store_in_model is used because the weekly seasonality coefficient will be computed during training and applied during prediction.

    calculated_cols:
  month_coefficient:
    method: seasonality
    input_columns:
    - qty_sold
    - month
    output_columns:
    - month_season_coeffficient
    store_in_model: true
    stored_keys:
    - month
    - qty_sold
    stored_columns:
    - month_season_coeffficient
    params:
      time_col: month
      value: qty_sold

Example of output:

month
month_season_coeffficient

1

0.95168072905404733

2

1.0357862750651963

3

1.0792659328597893

4

1.0395920562861072

5

1.0606537886934222

6

1.077858133811912

7

0.90191046323711832

8

0.95522180681369606

9

1.0252685243518904

10

0.97325249153619431

11

0.82898384742434672

12

0.99689396468790548

  1. Imagine you have model_resolution in your model. In this case, your coefficient will be computed per resolution. If you want to maintain the computation as described above, use global: true.

    preprocessing:  
  model_resolution:
  - country
  - pos_size
  
calculated_cols:
  month_coefficient:
    method: seasonality
    input_columns:
    - qty_sold
    - month
    output_columns:
    - month_season_coeffficient
    store_in_model: true
    stored_keys:
    - month
    - qty_sold
    stored_columns:
    - month_season_coeffficient
    params:
      time_col: month
      value: qty_sold

Example of output:

month
month_season_coeffficient
country
pos_size

1

0.56073028043659023

BELGIUM

JUMBO

2

0.7477214898040766

BELGIUM

JUMBO

1

0.97832471421812339

BELGIUM

MEDIUM

2

1.0817738908560632

BELGIUM

MEDIUM

1

0.8893741423322854

BELGIUM

XS

2

1.2777182380138707

BELGIUM

XS

1

0.94944234555496687

IRELAND

JUMBO

2

0.99525831058681336

IRELAND

JUMBO

1

0.927610409362476

IRELAND

LARGE

2

1.016215614376581

IRELAND

LARGE

1

0.89273852338090465

IRELAND

MEGA

2

1.0118941815702633

IRELAND

MEGA

1

0.9948854555639558

LUXEMBOURG

JUMBO

2

1.0520470613380593

LUXEMBOURG

JUMBO

1

0.93302404196451849

LUXEMBOURG

LARGE

2

0.95808824169437257

LUXEMBOURG

LARGE

Previousevents_countdownNexttsfresh

Last updated