-
Hi @AlexeyAB ,
Today, when I was browsing the web, I was surprised and happy to find your YOLOv3-spp.
What are the differences between YOLOv3-spp and YOLOv3? can you give a detailed
introduc…
-
Dear @AlexeyAB ,
Sorry for bothering you.
When I compare the Yolov3.cfg and Yolov3-spp.cfg, I find that you use spp structure to replace the
78 layer of Yolov3.cfg for creating Yolov3-spp.cfg.…
-
https://github.com/Tencent/ncnn/blob/e16b338b136c94805bc7d0ef3756f2dc4bfa3408/docs/how-to-use-and-FAQ/use-ncnn-with-pytorch-or-onnx.md
测试环境:
• OS: win10
• pytorch==1.4.0
• 编译工具:x64 Native Tools…
-
Thank you for your great work!
I trained ultralytics/yolov3 :yolov3-tiny、yolov3 and yolov3-spp.
I succeeded in training yolov3-tiny/yolov3 with one class on my own data and tracking the objects.
…
-
@AlexeyAB Thank you for your work, I encountered a problem
When I converted to the tensorrt model, I found that the darknet maxpool layer implementation is l-> out_w = (w + l-> pad-l-> size) / l-> st…
ou525 updated
4 years ago
-
Hi @AlexeyAB,
Yolov3-spp performs better on the Coco dataset than Yolov3.
When training with my on images and classes, do you recommend using the Yolov3-spp architecture as well? If so, can I us…
-
Hi,
I am just curious that will you publish yolov3-spp tiny version? If there is no in the plan, then how can we achieve from your current yolov3-spp architecture? Some hints might be good.
Tha…
-
-
想问一下,目前项目支持yolov3和yolov3-spp。如何修改使用yolov3-spp-pan-scale.cfg? 基础训练和稀疏训练可以支持yolov3-spp-pan-scale.cfg,在层和通道剪枝运行layer_channel_prune.py 时AttributeError: 'NoneType' object has no attribute 'reshape',如何修改这部分…
-
When I convert yolov3-spp to onnx, it reported:
` File "/home/lgy/PycharmProjects/TensorRT_yolo3_module/spp_weight_to_onnx.py", line 296, in _load_one_param_type
buffer=self.weights_file.read(p…