Start Here

We know you're eager to start labeling your data so we made this quick start guide that covers the basics! For more in-depth feature guides explore the topics on the left, or use the search bar above.

Questions? Contact us at [email protected]

1. Create your account and log in

  1. If it's your first time logging into Plainsight, sign into your account at or create an account.

  2. Before logging in, be sure to verify your email address by clicking on the link in the confirmation email.

Private cloud users will log in with a provided sign-in link. Please contact support if you have completed the deployment process but have not received your link.

2. Create a Dataset & Add Sources

In the main nav, you will be taken to the Datasets tab. This is where the labeler features can be found.

From here you can create a dataset, or add other users in the User Management settings.

  1. Click "Create New" to add a new dataset

  2. Give the dataset a name

  3. Click "Save & Continue to Sources"

  4. Select which type of datasource to add and fill in the setup details. Repeat for each source.

  5. When finished, click "Save & Continue to Label Definitions"

3. Create Label Definitions

Define the types of labels you will use in your dataset.

  1. Type label name in "Label Name" field

  2. Select label type from "Type" dropdown

  3. Select label color from color chooser (if applicable)

  4. Click "Save" button

  5. Repeat as needed for multiple labels

When you're finished, scroll down and click "Save & Start Labeling"

4. Label Images

  1. Select desired label from panel

  2. Apply Label to Image (Application will vary by label type)

  3. Click "Submit" button

  4. Repeat for each image

See Labeling Data for more about the labeler navigation & shortcuts.

5. Review, then Export or Train

Review your labels

  1. Click the "Review" tab

  2. Approve or reject images in the list

Train a model in Plainsight

  1. Go to the "Versions" tab

  2. Lock your dataset. If there are any unapproved labels, you will be prompted to auto-approve all.

  3. Select the desired model options for your training run:

    • Dataset and version. Only locked versions can be used.

    • Labels to be used in training. A model is trained on only one label type.

    • Training length, between 1-24 hrs. This is the amount of time spent training your model. Read more about Training Time.

  4. Click "Save & Start Training"

Only Rectangles (Bounding Box), Polygons (Instance Segmentation), and Feature Point (Keypoint) label types are currently supported in SmartML training. Other label types can be exported and trained outside of Plainsight.

Your model will begin training shortly. You will be notified by email or in-app notification when it is ready.

Export your annotations for training outside of Plainsight

  1. Go to the "Versions" tab

  2. Lock your dataset. If there are any unapproved labels, you will be prompted to auto-approve all.

  3. Select desired format for export

  4. Select or de-select desired label types to export

  5. Click "Export Now"

You will be notified by email and in-app notification when your export is ready. The labeled data will be available in a zip file for download.

Your data is now ready to use in your own project!