Skip to content

Deep learning solutions built on neural networks for computer vision applications - image identification

Notifications You must be signed in to change notification settings

chibuikeeugene/computer_vision_model_apis

Repository files navigation

COMPUTER VISION MODEL APIS

Deep learning applications built on neural networks for various business applications.

Description

This API service contains several packages built and hosted independently and finally imported into this project. There are three endpoints

Service endpoint 1 (Chest xray classifier)

  • returns prediction for condition(either normal or pneumonia) of a patient's chest health status

Service endpoint 2 (Kitchen ware classifier)

  • object identification - returns the name of the image of any of the trained objects as seen during development (glass, cup, spoon, plate, knife, fork)

Service endpoint 3 (Brain Scan image classifier)

  • returns prediction for condition(either glioma, meningioma, no tumor, or pituitary) of a patient's brain scan outcome

Dependencies and packages

  1. numpy>=1.20.0,<1.21.0
  2. python = 3.10
  3. pandas = 2.2.2
  4. scikit-learn = 1.4.2
  5. pydantic = 2.7.0
  6. strictyaml = 1.7.3
  7. tensorflow = 2.16.1
  8. scikeras = 0.13.0
  9. tensorflow-datasets = 4.9.4
  10. pillow = 10.3.0
  11. pydantic-settings = 2.2.1
  12. fastapi = 0.110.3
  13. uvicorn = 0.29.0
  14. loguru = 0.7.2
  15. python-multipart = 0.0.9
  16. kitchenware-model-package = 0.0.3
  17. braintumor-model-package = 0.1.0
  18. pneumonia_model_package = 0.2.0

Source code link

Source code link: Github link

Python index packages:

Snapshots

Api documentation 1

Api documentation 2

About

Deep learning solutions built on neural networks for computer vision applications - image identification

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published