Home

Upload from S3

When you import data from the S3 bucket, the data is added to the task queue automatically.

Note: You will not be able to select the images in the S3 bucket or folder to be added to task queue from our platform.

Steps to get your S3 upload details

  1. Log in to AWS.
  2. Click into Security Credentials from the top right dropdown
  3. Under 'Access keys for CLI, SDK, & API access', click Create access key
  4. Your keys will look something like this:
  • Access key ID example: AKIAIOSFODNN7EXAMPLE
  • Secret access key example: wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
  1. Copy the access key ID and secret access key and paste them into the respective fields on Bolt.
  2. On AWS, select the bucket you plan to import the data from.
  3. Visit the "Permissions" tab

  1. Create a bucket policy that allows Bolt to access your S3 bucket. You may refer to this sample policy.
{
    "Version": "2012-10-17",
    "Id": "Policy1546558291129",
    "Statement": [
        {
            "Sid": "Stmt1546558287955",
            "Effect": "Allow",
            "Principal": {
                "AWS": "885807555562"
            },
            "Action": [
                "s3:ListBucket",
                "s3:GetObject"
            ],
            "Resource": [
                "arn:aws:s3:::bucketname/*",
                "arn:aws:s3:::bucketname"
            ]
        }
    ]
}

The ID "885807555562" is SUPA's S3 ID and allows us to access the files stated in the Resource bucket path. Replace the AWS bucket name with your S3 bucket containing the files to be annotated.

  1. Enter the bucket and folder names on Bolt.
  2. You're now ready to upload your data!

S3 Subfolders

Databolt S3 upload supports subfolders.

To upload images from a subfolder, specify the folder path in the Folder Name, i.e folder1/folder2/folder3

For the inner-most subfolder in the specified Folder Name, all the images in the folder and its subfolders will be uploaded. If the Folder Name field is left blank, all the images in the bucket will be uploaded.

Here is an example of how the folder paths can be specified:

Assuming the folder and file structure below:

Below are some example inputs of the file path, and the expected images that will be uploaded:

CaseFolder NameImages that will be uploaded
Case 1nofolder, f1, f2, f3a, f3b, f4, f5
Case 2Folder1f1, f2, f3a, f3b, f4, f5
Case 3Folder1/Folder2f2, f3a, f3b, f4, f5
Case 4Folder1/Folder2/Folder3f3a, f3b
Case 5Folder1/Folder2/Folder3/Folder4f4, f5

To copy the file path from S3 for the "Folder Name" field:

You may copy from the breadcrumb in S3 directly: samplefolder1/samplefolder2