PlanIQ Error Messages and Warnings

Options

Users may encounter error and warning messages in PlanIQ. The table below includes descriptions of the messages and details what should be done in order to address them, where applicable.

 

Message types:

  1. Errors will fail the relevant PlanIQ step. Users will have to follow the instructions in order to avoid the errors.
  2. Warnings will not fail the relevant PlanIQ step. Users will be able to proceed as planned.

 

Message

Type

Description

Comments

{Dataset Name} headers contain the following reserved words: {Used Reserved Words}. Please change module line item names.

Error

Some field names are reserved by Amazon Forecast and therefore not allowed (see the full list of reserved words here). Please change the field names accordingly.

{Dataset Name} refers to the name of the dataset.

 

{Used Reserved Words} refers to a list of words reserved by Amazon Forecast that were used by the user.

Data type of historical / related time series column {Attribute Name} has changed from {Expected Data Type} and now includes {Value}. Please revert data type change in the source module.

Error

One of the data columns contains a data type (e.g. Number) that was later changed to a different data type. Please revert to the original data type.

 

{Attribute Name} refers to one of the data columns in the dataset.

 

{Expected Data Type} refers to the expected data type.

 

{Value} refers to a sample value from the data column that doesn’t align with the expected data type.

The data has less than 300 historical value records and cannot be used by advanced forecast algorithms such as DeepAR+, CNN-QR, Anaplan AutoML . Please add more historical data.

Warning

PlanIQ will not be able to use advanced algorithms (DeepAR+, CNN-QR, Anaplan AutoML) if less than 300 historical values were provided across all time series.

 

Historical data contains records that cannot be converted to numerical values. Please check the data.

Error

The prediction target column in the historical data contains non-numeric values (e.g. String or Date). Please check the historical data and export action for invalid values.

 

 

Unable to identify the time format in the historical data.

Error

The historical data includes a date format that PlanIQ cannot identify. Please check the date format and verify it is PlanIQ-compatible.

 

 

The historical data does not contain all the required columns.

Error

The historical data must contain three columns: item unique identifier, time dimension and prediction target. If some of those columns are missing, it will cause an issue with the export action. Please examine the historical data module and export action to verify that the three columns are available.

 

Maximum number of rows in the data collection must be less than 1,000,000.

Error

The maximum number of records in the historical data is 1,000,000 records. Please reduce the number of records that the export action generates. If possible, a view with filter should be utilized in order to achieve higher level of control.

 

The data collection contains more than 100,000 individual items. Please reduce number of individual items to 100,000.

Error

The historical data module may contain up to 100,000 distinct unique item identifiers. Please consider breaking down the dataset and training less unique items per forecast model.

 

 

Maximum number of columns in the historical data is 3. Please reduce number of columns to 3 or less.

Error

The historical data contains more than 3 columns. Required columns are: item unique identifier, time dimension and prediction target.

 

 

The historical data contains too many missing values, more than 50%. Please check if you can fill missing values.

Error

More than 50% of the values in the historical data are missing. Please review the data and consider filling the missing values or forecasting at a different level of granularity.

 

The historical data time scale could not be found.

Warning

This warning is presented if any of the following conditions in the historical data exist

·         Some dates are missing

·         There are duplicate dates

·         Some dates not aligned with the time scale of the historical data set (for example, if weekly scale is specified and all weeks are represented by values for Mondays, and occasionally there is a record with date on Sunday)

 

Please examine the date format in the historical data module and export action.

 

The prediction target column is empty. Please check the source module and export action.

Error

The prediction target column in the historical data module has no data. Please examine the module and export action.

 

 

The related data does not include all the required columns.

Error

The related data must have a minimum of two columns: item unique identifier and time dimension. If at least one of those columns is missing, it indicates that there is an issue with the configuration of the export action, or that previously-available columns in the related data export action are now missing. Please examine the related data module and export action carefully.

 

The related data includes the prediction value column. Please remove the {Prediction Value} column.

Error

The related data contains the prediction value column. This might indicate that there is an issue with the export action. For instance, you may have erroneously used your historical data export action for the related data. Please examine the related data export action file format accordingly.

{Prediction Value} refers to the prediction value column.

Unable to identify the time format in related data.

Error

The related data includes a date format that PlanIQ cannot identify. Please check the date format and verify it is PlanIQ-compatible.

 

Maximum number of columns in the related data is 13. Please reduce number of columns to 13 or less.

Error

The related data contains more than 13 columns (not including the item unique identifier and time dimension). Please reduce the number of columns.

 

 

All the numeric and boolean columns in the related data are empty. Please check the source module and export action.

Error

All the numeric and Boolean columns in the related data set (apart from item unique identifier and time dimension column) are empty. Please configure the export action properly.

 

Columns {Empty Columns} in the related data are empty. Please check the source module and export action.

Error

The related data columns listed in the error message are empty. Please validate that the export action is properly configured.

 

{Empty Columns} refers to columns in the related data set that are empty.

Related data time scale could not be found.

Warning

This warning is presented if any of the following conditions in the related data exist

·         Some dates are missing

·         There are duplicate dates

·         Some dates not aligned with the time scale of the historical data set (for example, if weekly scale is specified and all weeks are represented by values for Mondays, and occasionally there is a record with date on Sunday)

 

Please examine the date format in the related data module and export action.

 

There are more than 10 columns in the attributes dataset. Please reduce number of columns to 10.

Error

The attributes data contains more than 10 columns (not including the item unique identifier).  Please reduce the number of columns.

 

 

The attributes don't include the item id: {Item ID column}

Error

The attributes data does not include item unique identifier column. Please validate that the export action is properly configured.

{Item ID} refers to the item unique identifier column.

The attributes don't include all items in the historical data. Missing item ids: {Missing Items}

Error

Some items that appear in historical data are not present in the attributes data. All listed items must be present in the attributes data.

{Missing Items} refers to items that are present in the historical data but are missing in the attributes data.

Timeline of historical and related data does not match for items: {Invalid Items}. Related data timeline should start on or before historical data timeline.

Error

The related data must start on or before the start date of the historical data. This error message is displayed if at least one item’s start date in the historical data precedes its first appearance in the related data. Please adjust the data of the problematic items.

 

{Invalid Items} refers to items for which the timeline of the historical and related data doesn’t match.

The related and historical data do not contain the same item ids.

Warning

There is at least one item that exists in the historical data set but does not exist in the related data set, or vice versa.

 

The related data does not include forward looking information. Add forward looking data to improve forecast accuracy.

Warning

The related data does not include forward-looking information. Even though some algorithms can operate without forward-looking related data, it is advised to provide this data to improve forecast accuracy.

 

 

The time scale of related and historical data is incompatible.

Error

There are certain restrictions concerning the time scale of the related and historical data. Please refer to the relevant article for more details.

 

PlanIQ has been unable to check the model prediction quality based on the actuals that have been provided.

 

Warning

The model prediction quality comparison process cannot run because unified actuals timeline cannot be constructed.

Forecast action execution and generation of forecasts is not affected by this warning.

 

PlanIQ has been unable to check the model prediction quality based on the previous forecasts.

 

Warning

The model prediction quality comparison process cannot run because unified historic forecast timeline cannot be constructed.

Forecast action execution and generation of forecasts is not affected by this warning.

 

PlanIQ has been unable to check the model prediction quality.

 

Warning

The model prediction quality comparison process cannot run due to some internal error.

Forecast action execution and generation of forecasts is not affected by this warning.

 

Your prediction quality has reduced from {Original Predictor} to {Current Predictor}.

 

Warning

Model quality has deteriorated from either high → medium or low, or from medium → low.

 

Forecast action execution and generation of forecasts is not affected by this warning.

{Original Predictor} refers to the previous model quality.

 

{Current Predictor} refers to the current model quality.