AWS Sig v4 authorization example

Prerequisites

• Python 3.6 or higher

Install

$ pip install -r requirements.txt

Usage

You will need to acquire below values by contacting Lucidtech

• API key (<api_key>)
• AWS Access Key Id (<access_key_id>)
• AWS Secret Access Key (<secret_access_key>)
• API Endpoint Prefix (<endpoint_prefix>)

Note: Your REST API endpoint might not have support for all prediction types

Invoice prediction:

$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> invoice_prediction invoice.pdf application/pdf
$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> invoice_prediction invoice.jpeg image/jpeg

Receipt prediction:

$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> receipt_prediction receipt.pdf application/pdf
$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> receipt_prediction receipt.jpeg image/jpeg

Document split:

$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> document_split document.pdf application/pdf

Custom S3 KMS key

For APIs that encrypt S3 data with a custom KMS key instead of the default S3 key you need to provide the following flag to the program.

--with_s3_kms

An example

$ python src/main.py https://<endpoint_prefix>.api.lucidtech.ai/v1 <api_key> <access_key_id> <secret_access_key> --with_s3_kms invoice_prediction invoice.pdf application/pdf