AXERA-TECH / yolov5-qatLinks
Try to export the ONNX QDQ model that conforms to the AXERA NPU quantization specification. Currently, only w8a8 is supported.
☆11Updated last year
Alternatives and similar repositories for yolov5-qat
Users that are interested in yolov5-qat are comparing it to the libraries listed below
Sorting:
- ☆17Updated last year
- ☆23Updated 2 years ago
- 搜藏的希望的代码片段☆13Updated 2 years ago
- Llama3 Streaming Chat Sample☆22Updated last year
- Speed up image preprocess with cuda when handle image or tensorrt inference☆83Updated 2 months ago
- An onnx-based quantitation tool.☆71Updated 2 years ago
- Quantize yolov5 using pytorch_quantization.🚀🚀🚀☆14Updated 2 years ago
- 跟着Tensorrt_pro学习各种知识☆40Updated 3 years ago
- c++实现的clip推理,模型有一点点改动,但是不大,改动和导出模型的代码可以在readme里找到,模型文件都在Releases里,包括AX650的模型。新增支持ChineseCLIP☆30Updated 6 months ago
- 该代码与B站上的视频 https://www.bilibili.com/video/BV18L41197Uz/?spm_id_from=333.788&vd_source=eefa4b6e337f16d87d87c2c357db8ca7 相关联。☆71Updated 2 years ago
- Python scripts performing Open Vocabulary Object Detection using the YOLO-World model in ONNX. And Export the ONNX model for AXera's NPU☆12Updated 5 months ago
- Quantize yolov7 using pytorch_quantization.🚀🚀🚀☆11Updated 2 years ago
- ☆30Updated last year
- 对 tensorRT_Pro 开源项目理解☆22Updated 2 years ago
- ☆17Updated 2 years ago
- ☆47Updated 2 years ago
- ☆19Updated 3 years ago
- ☆22Updated last year
- ffmpeg+cuvid+tensorrt+multicamera☆12Updated last year
- ☆10Updated last year
- Using pattern matcher in onnx model to match and replace subgraphs.☆81Updated last year
- This is 8-bit quantization sample for yolov5. Both PTQ, QAT and Partial Quantization have been implemented, and present the results based…☆110Updated 3 years ago
- A set of examples around MegEngine☆31Updated 2 years ago
- DETR tensor去除推理过程无用辅助头+fp16部署再次加速+解决转tensorrt 输出全为0问题的新方法。☆12Updated 2 years ago
- A simple tutorial of SNPE.☆183Updated 2 years ago
- a simple pipline of int8 quantization based on tensorrt.☆69Updated 3 years ago
- ☆26Updated last year
- 高效部署:YOLO X, V3, V4, V5, V6, V7, V8, EdgeYOLO TRT推理 ™️ ,前后处理均由CUDA核函数实现 CPP/CUDA🚀☆53Updated 2 years ago
- ☆20Updated 2 years ago
- yolov8seg 瑞芯微 rknn 板端 C++部署,使用平台 rk3588。☆29Updated last year