This logic contains some functions needed specifically for this Accelerator, such as reading its configuration from the application settings, applying the user filters in each part of the model, preprocessing the data for the charts, and many small helpers for the charts rendering. The elements ParametersUtils, TablesUtils, and LabelsUtils contain the names of many elements, tables and fields of the models.

This lib is the place to change input names inside the model to reflect the user business vocabulary. LabelsUtils contains text that is shown in the UI, ParametersUtils and TablesUtilscontain table names and parameter names, respectively. Here you can also write functions to be used in different places of the model class.

This logic generates a parameter table containing default model training parameters. If the table already exists, it does nothing.

The TrainingParameters parameter table.

To change the default values in the table.

These logics define the Data Source and the mapping of the entries for the transactions, as well as some configuration of the outputs in the configurator. The main logic calls the configurator and the code for the dashboard portlets.

Two portlets show the selected data fields for the in-scope and filtered-out transactions.

The mapping could be changed if a field is removed or added. Default values of the configuration could be changed.

This logic defines more advanced configuration of the forecasting model.

The inputs are stored to be used by the Python job which trains the model.

Default values of the configuration could be changed.

These logics define the Additional Sources and the mapping of their fields to the fields of the source defined in the Definition step. The main logic calls the configurator and the code for the dashboard portlets.

The user inputs are stored to be used to load the sources into model tables in the Model Training Step. The contents of the selected sources are displayed in portlets.

Default values of the configuration could be changed.

The aim of this logic is to gather the data and convert it into a form that can be used to train the forecasting model, store the data, train the model and store the results tables.

The steps are as follows:

• StoreFraction.groovy – This element calculates the fraction of sales for each product in each store for the user inputted store_fraction_window and saves it as a table.

• DataQuery.groovy – This logic takes the mapping and configuration from the Definition tab and defines a query on the source data that converts it from transactions to a complete aggregated time series for each product, filling missing quantity data with 0 and missing price data with the last known value. The query is then loaded into the Data model table.

• DropExpiredTables.groovy – Drops the optional tables defined in the Additional Sources step to prevent them from being kept if the sources are removed.

• LoadEventsTable.groovy, LoadMappedAdditionalSourceTable.groovy, LoadMultiFeatureAdditionalSourceTable.groovy – These logics take the optional sources from the Additional Sources step and aggregate them to match the aggregation done to the source data in DataQuery.groovy before loading them into model tables.

• Train.groovy – This element collects the configuration for the forecasting model and triggers the Python job.

• TrainScript.py – This element is a Python script that loads the data, further processes them to add new feature columns based on the Model Configuration tab inputs and trains the machine learning model.

The Groovy elements output the store_quantity_fraction, and Data tables.

The Python job triggered in Train.groovy will write a set of model tables, including both the processed input data and the model results. It will also add a copy of the model as an attachment in a pickle file called model.pkl. This writing is done directly by the Python job and is not related to a Groovy logic.

The following tables are produced:

• processed_data – Final input data after all preprocessing.

• X_train, X_test, y_train, and y_test – Data split into training and test data.

• train_preds and test_preds – Training and test set model predictions, respectively.

• The tables with suffix _curve – Metrics data tables used for producing charts in the Model Training step.

• model_parameters – Model parameters used for training.

• shap_values – Data for the Feature Importances chart in the Model Training step.

• metrics_table – Data for the Metrics table in the Model Training step.

• events_encoded – If an events source is set in the Additional Sources step, the events are encoded into a form usable in model training and saved into this table.

If you wish to change the model to use the per unit item price instead of the extended price,
then change the following line in DataQuery.groovy:

(SUM($price) / NULLIF(SUM($quantity),0)) as price,
to
ISNULL((SUM($price * $quantity) / NULLIF(SUM($quantity), 0)), AVG($price)) as price,

This change will do two things: use the per unit item price instead of extended price to calculate the weighted average price, and default to the average price for a given time period if the total sales quantity in that time is 0.

This logic produces a dashboard with 3 charts and a table to visualize the model results.
PredictVsActualTrain.groovy and PredictVsActualTest.groovy produce two similar charts for the model training data and test data, respectively.

Four portlets. Three charts showing the feature importances and scatter plots of the model predictions vs. the historical quantity for both the training and test data, and a table summarizing performance statistics.

Add, modify or remove visualizations.

This logic produces two similar charts, one for a single product, the other for the sum of all products. The configurator selects a single product from the list of unique products in the Data table, which is used in singleProductForecast.groovy to display that products forecast.

Two portlets showing the demand forecast for a single product and for the sum of all products. The shaded area in green represents results from the test data, the remaining data is from the training set.

Add, modify or remove visualizations.

This logic displays the Training curve data. These are technical charts for assessing the model training.

Three portlets displaying the training curves for the three different metrics in the three _curve tables.

To modify or remove visualizations.

This logic displays a configurator allowing the user to choose the settings for elasticity calculation in the Model Predictions step.

All the user inputs of this configurator are available in the next step.

Default values of the configuration could be changed.

This calculation consists of the following steps:

• FallbackProductsMinTransactions.groovy – If a fallback level is set, this element calculates the list of products which do not meet the minimum transaction threshold set in the Elasticity Settings tab of the Model Training step. The list of products is loaded into the fallback_products model table.

• DropExpiredElasticityTables.groovy – This element drops expired elasticity tables if the “Calculate Elasticities” checkbox is unticked in the elasticity settings tab.

• DropExpiredFallbackTable.groovy – This element drops expired fallback tables if the fallback level is not set in the Elasticity Settings tab.

• Elasticity.groovy – This element collects the configuration for the elasticity calculation and triggers the Python job.

• ElasticityScript.py – This element is a Python script that loads the data and then trains a final machine learning model using the parameters in model_parameters on the full dataset, producing a future forecast. It then uses the predictions from this model to generate elasticity curves for each product and number of time periods of the forecast and fits a logistic ('S') curve to the data.

The Python job triggered in Elasticity.groovy will write a set of model tables. This writing is done directly by the Python job and is not related to a Groovy logic.

The following tables are produced:

• forecast – Future forecast outputs.

• final_model_preds – Predictions of the final model on the training data.

• elasticity_factors – Fitted curve parameters needed to reproduce the elasticity curves.

• elasticity_data – Data used for plotting the elasticity curve charts.

• elasticity_r2_scores – r² scores for the logistic curve fits.

• simple_elasticity – The simple elasticity data to be used as the fallback for products that do not meet the minimum requirements chosen in the Elasticity Settings tab.

• fallback_products – The products that do not meet the minimum transaction and minimum r² score requirements chosen in the Elasticity Settings tab.

There is no reason to edit this in general.

This calculation precomputes the data needed for the summed forecast of all products portlet and stores it in a table. It is then loaded in the Overview Tab, leading to faster rendering and avoiding potential timeouts with large datasets.

Table summed_forecast used in Overview Tab.

There is no reason to edit this in general.

This logic displays charts and tables for the general results of the elasticity calculation for all products.

Portlets showing:

• Summed forecast for all products.

• Several histograms showing the distribution of the elasticity metrics.

• Tables showing the elasticity and fallback elasticity data.

To add, modify, or remove charts or tables.

This logic displays charts and a table showing the results of the elasticity calculation for a single product and date of the forecast. The configurator selects a single product from the list of unique products in the forecast table, and a date from the available dates in the forecast table.

Portlets showing:

• Forecast for a single product.

• Elasticity curve for the product and period start date selected in the configurator.

• Table showing the elasticity_factors data for the selected product.

To add, modify, or remove charts or tables.

It is a way for the models to communicate outside of themselves some basic information about their state:

• It can provide the names of certain output tables so they may be accessed outside of the model.

• It provides some model configuration.

The usage to request the table names is:

This will output a map with the following entries:

• ElasticityFactorsTableName - a string. The table name for the elasticity curve parameter data (elasticity_factorsby default).

• ElasticityFallbackTableName - a string. The table name for the fallback elasticities (simple_elasticity by default).

The usage to request the configuration is:

This will output a map of the following configuration values:

• FORECAST_LENGTH - an integer.

• STORE_FRACTION_WINDOW - an integer.

• PERIOD - a string, either ‘daily’, ‘weekly’, or 'monthly.

• MAX_FORECAST_DATE - a date. The final date in the forecast.

• FALLBACK_LEVEL - a string. The aggregaton level used for the fallback.

To add more configuration outputs or add more output tables.