in scripts/comprehend_create_custom_entity.py [0:0]
def lambda_handler(event, context):
#####################Comprehend create Custom entity #################################\
######## 1- creat Lambda function Author from scratch with runtime Python 3.8
########2- Use Lambda-S3-Glue-comprehend role
########3- Increase Timeout to 2 min
#Note that the S3 bucket must be in the same region of Comprehend.
import boto3
import uuid
account_id = boto3.client("sts").get_caller_identity()["Account"]
comprehend = boto3.client("comprehend")
response = comprehend.create_entity_recognizer(
RecognizerName="Recognizer-blog".format(str(uuid.uuid4())),
LanguageCode="en",
DataAccessRoleArn = "arn:aws:iam::"+account_id+":role/Lambda-S3-Glue-comprehend",
InputDataConfig={
'DataFormat': 'COMPREHEND_CSV',
"EntityTypes": [
{"Type": "customer information ID",
},
{"Type": "current level of education",
},
{"Type": "customer credit rating",
}
],
'Documents': {
'S3Uri': 's3://<Your S3 Bucket>/traning/training_dataset.csv'},
'EntityList': {
'S3Uri': 's3://<Your S3 Bucket>/traning/labels_entity_list.csv'},
} )