Step-1: Create a policy and execution role for lambda to access DynamoDB.
You can do this while creating the lambda function. You will be required to select a simple microservice permission policy template.

Step 2: Create a lambda function with the below code.
You can do this by copying this code to the lambda_function.py and then deploy the code.

import boto3
import json

# define the DynamoDB table that Lambda will connect to
tableName = "lambda-apigateway"

# create the DynamoDB resource
dynamo = boto3.resource('dynamodb').Table(tableName)

print('Loading function')

def lambda_handler(event, context):
    '''Provide an event that contains the following keys:

      - operation: one of the operations in the operations dict below
      - payload: a JSON object containing parameters to pass to the 
                 operation being performed
    '''

    # define the functions used to perform the CRUD operations
    def ddb_update(x):
        dynamo.update_item(**x)

    def echo(x):
        return x

    operation = event['operation']

    operations = {
        'update': ddb_update,
        'echo': echo,
    }

    if operation in operations:
        return operations[operation](event.get('payload'))
    else:
        raise ValueError('Unrecognized operation "{}"'.format(operation))

Step-3: Create a REST API through api gateway.
First, you need to create a resource register. After creating a resource you need to create a post method. After creating a post create a new stage and deploy.

Step-4: Create a dynamodb to store user data.

Please create a dynamodb with lambda-API gateway as the table name and id as the partition key as shown below.

Step-5: After dynamo db is created use the below input to create an entry in dynamodb to test the function. by using the test function of api gateway.

{
    "operation": "update",
    "payload": {
        "Key": {
            "id": "1234ABCD"
        },
        "AttributeUpdates": {
            "number": {
                "Value": 10
            }
        }
    }
}