Importing data & annotations

You can programmatically import data and/or annotations using the SDK with a Python Script. For simple import operations, we recommend using the CLI, which has a simple & optimized interface. 
You can import either locally stored data or externally stored data using the SDK using create_datapoints.
Please see the full reference documentation for create_datapoints here.
Perform the standard RedBrick AI SDK set-up to create a project object.
project = redbrick.get_project(org_id, project_id, api_key)
Import locally stored data
To import locally stored data, create a list of points with relative file paths to your locally stored data, and use the redbrick.StorageMethod.REDBRICK storage ID. 
points = [{"items": ["path/to/data.nii"], "name": "..."}]
storage_id = redbrick.StorageMethod.REDBRICK
​
project.upload.create_datapoints(storage_id=storage_id, points=points)
Please visit the Items List documentation to understand the format of the points array.
Import externally stored data
To import data stored in an external storage method, like AWS s3, make sure to use the storage methods Storage ID found on the Storage tab of your RedBrick AI account.
Import annotations
If you want to upload annotations with your data, include the annotation information in the points object.
points = [
    {
        "name": "...",
        "series": [
            {
                "items": "path/to/data.nii",
                "segmentation": "path/to/segmentation.nii",
                "segmentMap": {1: "category-1", 2: "category-2"},
            }
        ],
    }
]
project.upload.create_datapoints(storage_id="...", points=points)
Please visit our annotation format reference to understand the format of the annotations.
Exporting tasks
Programmatic Label & Review
Last modified 2mo ago