Metadata-Version: 2.1
Name: sahi-fei
Version: 0.11.19
Summary: A vision library for performing sliced inference on large images/small objects
Home-page: https://github.com/glock-fei/sahi
Author: glock-fei
License: MIT
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Topic :: Software Development :: Libraries
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: opencv-python<=4.9.0.80
Requires-Dist: shapely>=1.8.0
Requires-Dist: tqdm>=4.48.2
Requires-Dist: pillow>=8.2.0
Requires-Dist: pybboxes==0.1.6
Requires-Dist: pyyaml
Requires-Dist: fire
Requires-Dist: terminaltables
Requires-Dist: requests
Requires-Dist: click
Requires-Dist: numpy<2.0.0
Provides-Extra: tests
Requires-Dist: mmdet==3.0.0; extra == "tests"
Requires-Dist: pycocotools==2.0.6; extra == "tests"
Provides-Extra: dev
Requires-Dist: black==22.3.0; extra == "dev"
Requires-Dist: flake8==3.9.2; extra == "dev"
Requires-Dist: importlib-metadata<4.3,>=1.1.0; python_version < "3.8" and extra == "dev"
Requires-Dist: isort==5.9.2; extra == "dev"
Requires-Dist: jupyterlab==3.0.14; extra == "dev"

## 切片推理

此代码 Fork [obss/sahi](https://github.com/obss/sahi), 添加如下功能
- 添加rknn支持
```python
detection_model = AutoDetectionModel.from_pretrained(
    model_type='yolov8rknn',
    model_path=yolov8_onnx_model_path,
    confidence_threshold=0.3,
    category_mapping=category_mapping,
    device="cpu", # or 'cuda:0'
)
```
- 添加yolov10 导出 onnx 支持
```python
detection_model = AutoDetectionModel.from_pretrained(
    model_type='yolov10onnx',
    model_path=yolov8_onnx_model_path,
    confidence_threshold=0.3,
    category_mapping=category_mapping,
)
```
- 添加参数 EP_LIST, 使用onnx推理时直接传递
```python
detection_model = AutoDetectionModel.from_pretrained(
    model_type='yolov10onnx',
    model_path=yolov8_onnx_model_path,
    confidence_threshold=0.3,
    category_mapping=category_mapping,
    ep_list=['CUDAExecutionProvider', 'CPUExecutionProvider']
)
```
