xunkaixin

Results 6 issues of xunkaixin

Hi, thanks, good work when pruning, some problems: return surgeon.operate() File "C:\Users\qkj\Anaconda3\lib\site-packages\kerassurgeon\surgeon.py", line 158, in operate self._mod_func_map[node](node, outputs, output_masks, **kwargs) File "C:\Users\qkj\Anaconda3\lib\site-packages\kerassurgeon\surgeon.py", line 309, in _delete_channels new_delete_mask = self._make_delete_mask(old_layer, channels)...

Model : yolov3 Backbone : efficientnet Questions : different scale width large different loss (train from scratch) Log files : I0909 14:54:48.172024 25166 solver.cpp:253] Iteration 418500 (143035 iter/s, 2.92586s/100 iters),...

Hello I tried to run the code test_segmentation.cpp CPU_ONLY . I get the error ./test_segmentation.bin Example_Models/bayesian_segnet_camvid.prototxt weights/bayesian_segnet_camvid.caffemodel CamVid/test/Seq05VD_f01230.png Scripts/camvid12.png [libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 450:24: Message type "caffe.DropoutParameter"...

'draw_net.py' can not draw the 'train_posenet.prototxt' net . the result is a blank image.

[objs.pcd.zip](https://github.com/user-attachments/files/16064708/objs.pcd.zip)

Hello, thanks a lot for your great work on this repository! I am training a small-object detector based on your configuration pretrain_r50_deformable_detr_dancetrack.yaml, using the pretrained model r50_deformable_detr_coco.pth as initialization. However,...