edgeai-yolov5

edgeai-yolov5 copied to clipboard

Failed to export onnx model using half precision

1

## ❔Question Does this project support export script to export semi-precision onnx models? ## Additional context 

PYLuck

For the specified number of key points custom dataset debug（针对自定义关键点数量的yolopose的debug）

3

我总结了一下需要改的地方： https://zhuanlan.zhihu.com/p/603799078 如果对您有帮助给个点赞收藏哈哈哈

divided7

## ❔Question python .\detect.py --weights last.pt --source 0 --kpt-label edgeai-yolov5-yolo-pose\detect.py", line 65, in detect model(torch.zeros(1, 3, imgsz[0], imgsz[0]).to(device).type_as(next(model.parameters()))) # run once TypeError: 'int' object is not subscriptable ## Additional context...

GreenTeaHua

## ❔Question deploy onnx with c++, how to know the predict num, can know with python ## Additional context 1. the python can print the out shape session = onnxruntime.InferenceSession(model_path,...

XDxiaoqi

## ❔Question Dear @debapriyamaji, thanks a lot for providing the code of YOLOv5Pose. I was wondering, - Did you release the scripts (functions, classes, utilities) for quantization? - Could you...

escorciav

Hello community! Could someone provide a guide on how to make the pose model multiclass? What would need to be adjusted besides the architecture? Greetings, Nico

NicolasPetermann134

pretrained model is wrong

1

## ❔Question yolov5s6-pose pretrained model: http://software-dl.ti.com/jacinto7/esd/modelzoo/gplv3/08_02_00_11/edgeai-yolov5/pretrained_models/checkpoints/keypoint/coco/edgeai-yolov5/other/person_detector/yolov5s6_960_71p6_93p1/weights/last.pt but the weights in opt.yaml is /user/a0132471/Files/bit-bucket/pytorch/yolov5_pose/runs/train/shared_models/person_detector/yolov5s6_person_64p8_90p2/weights/last.pt I used the pretrained model to obtain map0.547, which is lower than the readme 0.607 ## Additional...

zcunyi

## ❔Question Hello. I am new to Jacinto Family. I am confused about its deployment process. Could you please give some resources about how to make binaries from ONNX for...

sharif-apu

Hi author, Thanks for your excellent work. For yolo pose, the default anchors in models/hub/yolov5s6_kpts_ti_lite.yaml look the same as "anchors_p6_1280" defined in models/hub/anchors.yaml. My question is, if I select 640...

liamsun2019

hello, when i use cuda to export onnx, some mistakes occur, my command is python export.py --weights pretrained_models/yolov5s6_640_ti_lite/weights/best.pt --img 640 --batch 1 --simplify --export-nms --opset 11 --device 0# export at...

Onehundred0906