How to pass S3 object and Key to HumanLoopInput of start_human_loop

Question

I am able to extract text from my multi-page PDF using Amazon Textract. Now I want start Human Loop Review. I have already created a workflow and specified the condition there to trigger the Human Loop. Below is my code: -

import os
import json
import time
import uuid
from urllib.parse import unquote_plus
import boto3

def lambda_handler(event, context):
    textract = boto3.client("textract")
    a2i = boto3.client("sagemaker-a2i-runtime")
    FLOW_ARN = os.environ["FLOW_ARN"]
    if event:
        file_obj = event["Records"][0]
        bucketname = str(file_obj["s3"]["bucket"]["name"])
        filename = unquote_plus(str(file_obj["s3"]["object"]["key"]))
        
        # Start document analysis for the whole document
        response = textract.start_document_analysis(
            DocumentLocation={
                "S3Object": {
                    "Bucket": bucketname,
                    "Name": filename,
                }
            },
            FeatureTypes=["FORMS"],  # Specify the feature types to analyze
            ClientRequestToken=str(uuid.uuid4()),  # Generate a unique client request token
        )
        
        # Retrieve the job ID from the response
        job_id = response["JobId"]
        
        # Poll for the completion of the job
        while True:
            job_status = textract.get_document_analysis(JobId=job_id)['JobStatus']
            if job_status in ['SUCCEEDED', 'FAILED']:
                break
            time.sleep(5)  # Wait for 5 seconds before checking again
        
        # Get the results of the analysis
        response = textract.get_document_analysis(JobId=job_id)
        
        # Process the results
        print(json.dumps(response))
        
        a2i.start_human_loop(
            HumanLoopName=uuid.uuid4().hex,
            FlowDefinitionArn=FLOW_ARN,
            HumanLoopInput={
                'InputContent': json.dumps({
                    "InitialValue": {
                        "Bucket": bucketname,
                            "DocumentPath": filename,
                         }
                    })
                },
                DataAttributes={
                    'ContentClassifiers': [
                        'FreeOfAdultContent',
                    ]
                }
            )

        return {
            "statusCode": 200,
            "body": json.dumps("Document processed successfully!"),
        }

    return {"statusCode": 500, "body": json.dumps("Issue processing file!")}

I was expecting it to start the human loop review but it return following error: -

[ERROR] ValidationException: An error occurred (ValidationException) when calling the StartHumanLoop operation: Provided InputContent is not valid. Please use valid InputContent JSON and try your request again.

Could someone please point what I am doing wrong? I need to pass my PDF in S3 bucket to HumanLoopInput.

--------------------EDIT------------------------------

I am using default worker template, here it is: -

<script src="https://assets.crowd.aws/crowd-html-elements.js"></script>
{% capture s3_uri %}s3://{{ task.input.aiServiceRequest.document.s3Object.bucket }}/{{ task.input.aiServiceRequest.document.s3Object.name }}{% endcapture %}

<crowd-form>
  <crowd-textract-analyze-document src="{{ s3_uri | grant_read_access }}" initial-value="{{ task.input.selectedAiServiceResponse.blocks }}" header="Review the key-value pairs listed on the right and correct them if they don't match the following document." no-key-edit="" no-geometry-edit="" keys="{{ task.input.humanLoopContext.importantFormKeys }}" block-types="['KEY_VALUE_SET']">
<short-instructions header="Instructions"><p>Click on a key-value block to highlight the corresponding key-value pair in the document.</p><p><br></p><p>If it is a valid key-value pair, review the content for the value. If the content is incorrect, correct it.</p><p><br></p><p>If the text of the value is incorrect, correct it.</p><p><img src="https://assets.crowd.aws/images/a2i-console/correct-value-text.png" width="100%"></p><p><br></p><p>If a wrong value is identified, correct it.</p><p><img src="https://assets.crowd.aws/images/a2i-console/correct-value.png" width="100%"></p><p><br></p><p>If it is not a valid key-value relationship, choose <strong>No</strong>.</p><p><img src="https://assets.crowd.aws/images/a2i-console/not-a-key-value-pair.png" width="100%"></p><p><br></p><p>If you can’t find the key in the document, choose <strong>Key not found</strong>.</p><p><img src="https://assets.crowd.aws/images/a2i-console/key-is-not-found.png" width="100%"></p><p><br></p><p>If the content of a field is empty, choose <strong>Value is blank</strong>.</p><p><img src="https://assets.crowd.aws/images/a2i-console/value-is-blank.png" width="100%"></p><p><br></p><p><strong>Examples</strong></p><p>The key and value are often displayed next or below to each other.</p><p><br></p><p>For example, key and value displayed in one line.</p><p><img src="https://assets.crowd.aws/images/a2i-console/sample-key-value-pair-1.png" width="100%"></p><p><br></p><p>For example, key and value displayed in two lines.</p><p><img src="https://assets.crowd.aws/images/a2i-console/sample-key-value-pair-2.png" width="100%"></p><p><br></p><p>If the content of the value has multiple lines, enter all the text without a line break. Include all value text, even if it extends beyond the highlighted box.</p><p><img src="https://assets.crowd.aws/images/a2i-console/multiple-lines.png" width="100%"></p></short-instructions>

<full-instructions header="Instructions"></full-instructions>
  </crowd-textract-analyze-document>
</crowd-form>

I can see below keys in this snippet: -

>     task.input.aiServiceRequest.document.s3Object.bucket
      task.input.aiServiceRequest.document.s3Object.name
      task.input.selectedAiServiceResponse.blocks
      task.input.humanLoopContext.importantFormKeys

But it looks like, it is calling some internal libraries of AWS, because it is a default template.

Answer 1

The keys in your InputContent need to be aligned with the human task UI template, which you have not shared.

Check out this example: https://github.com/aws-samples/amazon-textract-a2i-dynamodb-handwritten-tabular/blob/main/textract-hand-written-a2i-forms.ipynb

Notice the keys in InputContent are also found in the human task UI template. Are your keys aligned with your UI template?