Qishang Cheng

How to train the pvanet on imagenet?

2

I want to rebuild the pvanet from scratch,so I want to train the feature extraction network of pvanet on ImageNet2012,Can you give me some tips on training the network? @sanghoon

Will you create the soft_gpu_nms in gpu_nms.pyx?

7

I found if i use the soft_nms, i should set the USE_GPU_NMS: false. If I did not set the variable, your soft_nms does not work. I find there is just...

How to use pomodoro-cli on win7?

2

I have installed the program on my PC with win7, How should i use it?

Error with main.py

11

main.py --train --exp lr7e-3 --epochs 50 --base_lr 0.007 raise ValueError('Expected more than 1 value per channel when training, got input size {}'.format(size)) ValueError: Expected more than 1 value per channel...

ImportError: No module named dota_kit.poly_nms_gpu.poly_overlaps

11

 After running the following command. there are no poly_overlaps to import sudo apt-get install swig cd ${RoI_ROOT}/dota_kit swig -c++ -python polyiou.i python setup.py build_ext --inplace cd ${RoI_ROOT}/dota_kit/poly_nms_gpu make -j16

Segmentation Fault

18

I have run the six scripts in the examples folder. The arma.py, rgcn.py, sgc.py ere fine, however the cora.py, gat.py and gat.py were failed. Error: Segmentation fault (core dumped) System:...

I have reproduced the IOUloss by pytorch in mmdetection and replace the smmothl1 by it. However I got NaN. the code is below: def IOULoss(pred, target,weight, avg_factor=None): """ :param pred:...

SIGSEGV (@0x8) received by PID 33937 (TID 0x7fa21e15b800) from PID 8; stack trace

1

 when I put your range_loss.cpp and range_loss.hpp in my caffe, and modified the caffe.proto as the readme sayed, the compilation is sucessful. but when i run the code...

can you share your scripts of plotting the img?

1

I did not know how to plot this image, can you show me the method of plotting this kind of image? https://github.com/Charrin/RangeLoss-Caffe/blob/master/mnist/rl_st.png