- This meta project consists of simple script which can be used to create machine learning API in python which includes separate template for creating
preprocessing
,training
andprediction
API along withlogging
andbasic error
handling facility. - The preprocessing, training, prediction module consists of dummy code which you can fill as per your wish.
- The APIs are created in 2 different ways:
- Bare flask API
- Swagger based API
- Kindly look into the
simple_api/
directory here.
Swagger helps for API documentation which is quite helpful and a good practice.
- Kindly look into the
swagger_api/
directory here.
- Please look here
- Please look here
- Multiple model syndrome and how to avoid them while deploying ML model using REST API ⚡
- Also includes load-testing using Locust ⚡
- Add
requirement.txt
file - Implement with
fastapi
[click here]- Add swagger UI
- Add error code
- Add simple web UI
- Message Queue - Redis
- Load testing with Locust