The main role of the project:
- PyTorch's usage
- FiftyOne's usage
- ONNX's usage
- ONNX Runtime's usage
- OpenMMLab's usage
Note:
- support platform: linux, windows
- it is recommended to install Anaconda
- when executing the test program, you need to locate the terminal to the directory where the python file is currently executed
Test code:
- PyTorch code
- batch normalization
- Faster R-CNN + ResNet-50 target detection
- AlexNet image classification
- digital identification: LeNet-5(train and predict)
- ONNX/ONNX Runtime code
- ResNet-50 image classification
- LeNet-5 digital identification
- OpenMMLab code
- MMDetection
- MMEditing
- inpainting
- matting
- image super resolution
- image generation
- MMSegmentation
- MMClassification
- MMPose
- 2d face landmark
- 2d hand pose estimation
- 2d human pose estimation
- MMOCR
- MMGeneration
- MMDeploy
The version of each open source library see: version.txt
Blog: fengbingchun