Open xinsuinizhuan opened 2 years ago
is there the yolov7-tiny pretrained model?
could provide the yolov7-tiny.pt pretrained model?
OK, I will generate one soon.
can we use the cfg/deploy/yolov7-tiny-silu.yaml to train pytorch?
yes, but silu is not good for edge device, we do not recommend use silu in tiny model.
yes, but silu is not good for edge device, we do not recommend use silu in tiny model. can support the yolov7-tiny.yaml not silu for pytorch.
@xinsuinizhuan yolov7-tiny.pt
@xinsuinizhuan yolov7-tiny.pt
OK.Thank you. Let me have a try. VS yolov5s and yolov6s
the yolov7-tiny.pt include the silu?
No, it uses Leaky ReLU.
the cfg/deploy/yolov7-tiny.yaml file can release?
@wang911205 yolov7-tiny.yaml
the file is yolov7-tiny-silu.yaml include the silu
Do you have pretrained YOLOv7 with LeakyReLU? How many mAP drop comparing to SILU version?
ubuntu cuda 11.2 cudnn 8.1.1 1080ti
pt inference only: yolov5n: 5.5ms yolov7-tiny: 7. 5ms
windows vs 2019 cuda 11.3 cudnn 8.2.3 tensorrt 8.2 3060ti amd 5900H
tensorrt: yolov7-tiny: 3ms
Could you help for testing size 512, 416, and 320, thanks.
So no silu version is provided yet? The paper is reporting a 38.7% mAP (can this be released?), but I tested with the provided 'yolov7-tiny.pt' (actually leaky-relu version), got only 37.4%.
windows vs 2019 cuda 11.3 cudnn 8.2.3 tensorrt 8.2 3060ti amd 5900H
openvino forward: yolov5s: 90ms yolov6s 75, yolov7 282ms opencv dnn forward: yolov5s:21, yolov6s 18ms, yolov7 25ms tensorrt yolov5s: 6ms yolov6s 5ms yolov7 11ms