Historical Data
  • 10 Mar 2023
  • 2 Minutes to read
  • Dark
  • PDF

Historical Data

  • Dark
  • PDF

Article Summary

When you run inference using an API call, LandingLens automatically saves the image data to the Historical Data section on the Deploy page. In addition to reviewing results on this page, you can also refer back to the images later and use them to retrain your model.

The Historical Data Tab Displays Predicted Images
  • For legacy "classic" workflow users, the Historical Data section includes results from the Predict tool. To prevent Predict results from skewing your production data, newer versions of LandingLens don't include Predict results.
  • The images in this section are from an earlier version of LandingLens. Your interface might look a little different, but the basic features and tools are the same.

Save Predicted Images to Your Project

You can save Predicted Images to your Model to retrain that Model with those Images. For example, let's say that you have a Model trained to detect hardware in cereal, but the Model did not detect a screw in some Predicted Images. You can save these Predicted Images to your Model, label them, then retrain your Model. 

To save Predicted Images to your Model:

  1. Go to the Deploy page.
  2. Click the Historical Data section.
  3. Select the images you want to save to your Model.
  4. Click Add to Data Browser. The image will have the status In Sync.
    Add to Data Browser

Image Statuses

Images on the Historical Data tab can display two statuses: Raw and In Sync.

  • Raw means that the image is the original that was uploaded to the Project.
  • In Sync means that the image was sent to the Data Browser.
"Raw" and "In Sync" Statuses

Image Detail Settings

When you hover over an image in the Historical Data tab, the icon displays. Click this icon to view the image details.

Click "i" to View the Image Details


Image Details

The table below describes the sections on the Image Details pop-up window. 

1Toggle Full-Screen / Exit Full-Screen Enters and exits the full-screen mode.
2Image EnhancementManually adjust the brightness and contrast. Or choose an option to add enhancements automatically. These options are useful if an image is too dark and you want to brighten it so you can better see the details of the image, for example.
3Show LabelsToggle on and off the labels on the image. This is useful if you want to search the image for any missed items, for example.
4Directional KeysNavigate to the next or previous image.
5HotkeysView a list of all the available hotkeys (keyboard shortcuts).
6Close WindowClose the Image Details pop-up window.
7Confidence ScoreIf the Model detects items in an image, the Confidence Score displays. The Confidence Score represents how confident the Model is that its Prediction is correct. For example, in the screenshot above, the model is 19% (0.19) and 17% (0.17) confident that the items detected are screws.
8InformationIf you used an API to upload metadata, that metadata will display in this section. The Media ID is generated internally in the database.
9Human JudgmentData populates in this section after you or someone else marks a Prediction as "Correct" or "Incorrect".
  • Inspector ID: The user's email address who marked the Prediction.
  • Judgment: Displays OK if the prediction is correct or NG (Not Good) if the prediction is incorrect.
  • Comment: Displays any feedback left by the user who reviewed the Prediction.
10DownloadDownload the original image. This image will not show any Predictions. 
11Correct/IncorrectAllows users to mark Predictions as "Correct" or "Incorrect."

Was this article helpful?

What's Next