Download Raw Data

FactoryTalk Analytics GuardianAI allows the user to download the raw data of training, deviation, and failure risk to analyze or enable bulk training on a new algorithm.
  1. Navigate to the single component view page and click
    Download Data
    . A ZIP file containing the raw data for the component will be downloaded.
    If multiple states are available, the downloaded ZIP file includes the raw data for each of them.
Download Data
Following is the naming convention for the downloaded zip file:
<Drive Name>_RawData_<Time stamp>.zip
  • Drive Name: Name of the drive given during configuration.
  • Time Stamp: The exact time at which the data file was downloaded.
When user extracts the downloaded file, it contains a primary folder named using a unique identifier. If multiple states are included, it contains multiple folders—each corresponding to a state and labeled with its respective unique ID.
The data for each frequency bucket corresponding to a state will be stored in a separate folder. For example, data related to 48 Hz frequency will be stored in a folder named
48
. If a component is trained with multiple frequencies for a particular state, multiple folders will be created, each named after its respective frequency.
Each frequency folder contains the following data:
  • Training: The data for training is available as
    Training.zip
     file.
  • Deviation: The data for a deviation is available as
    <DeviationID>.zip
     inside a sub-folder named
    Deviation
    .
  • FailureRisk: The data for a FailureRisk is available as
    <FailureRiskID>.zip
     inside a sub-folder named
    FailureRisk
    .
Provide Feedback
Have questions or feedback about this documentation? Please submit your feedback here.
Normal