Metadata-Version: 2.1
Name: easy-inference
Version: 0.0.2
Summary: Got a working detection model file? Want to quickly setup inference pipelines? You are in the right place!
Home-page: UNKNOWN
Author: Chadi Salmi
License: MIT
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: numpy
Provides-Extra: gpu
Requires-Dist: pyopencl ; extra == 'gpu'
Requires-Dist: six ; extra == 'gpu'
Provides-Extra: testing
Requires-Dist: mypy ; extra == 'testing'
Requires-Dist: onnx ; extra == 'testing'
Requires-Dist: onnx2torch ; extra == 'testing'
Requires-Dist: protobuf (~=3.19.0) ; extra == 'testing'
Requires-Dist: pytest ; extra == 'testing'
Requires-Dist: torch (~=1.11.0) ; extra == 'testing'
Requires-Dist: tqdm ; extra == 'testing'

# Easy Inference

Welcome to the easy inference repository! The main goal of this repository is to provide a clean, simple and short way of setting up inference pipelines for 2D (and 3D) visual detection.
The interfaces to camera drivers are abstracted away as python `generators`. A simple inference pipeline for a webcam based inference pipeline looks as follows:

```Python3
from easy_inference.providers.webcam import Webcam

provider = Webcam(source=0)

for frame in provider:

  # run my detection 
```

See the examples directory for some `yolov7` pipelines.



