chuanqi305 / MobileNet-SSD

Caffe implementation of Google MobileNet SSD detection network, with pretrained weights on VOC0712 and mAP=0.727.
MIT License
2k stars 1.18k forks source link

when I run train.sh #179

Open PointCloudNiphon opened 4 years ago

PointCloudNiphon commented 4 years ago

`I0403 19:05:33.036474 257810 layer_factory.hpp:77] Creating layer data I0403 19:05:33.036964 257810 net.cpp:100] Creating Layer data I0403 19:05:33.036980 257810 net.cpp:408] data -> data I0403 19:05:33.037021 257810 net.cpp:408] data -> label I0403 19:05:33.041712 257836 db_lmdb.cpp:35] Opened lmdb trainval_lmdb/ I0403 19:05:33.065735 257810 annotated_data_layer.cpp:62] output data size: 24,3,300,300 I0403 19:05:33.109791 257810 net.cpp:150] Setting up data I0403 19:05:33.109824 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.109830 257810 net.cpp:157] Top shape: 1 1 1 8 (8) I0403 19:05:33.109836 257810 net.cpp:165] Memory required for data: 25920032 I0403 19:05:33.109853 257810 layer_factory.hpp:77] Creating layer data_data_0_split I0403 19:05:33.109870 257810 net.cpp:100] Creating Layer data_data_0_split I0403 19:05:33.109879 257810 net.cpp:434] data_data_0_split <- data I0403 19:05:33.109896 257810 net.cpp:408] data_data_0_split -> data_data_0_split_0 I0403 19:05:33.109911 257810 net.cpp:408] data_data_0_split -> data_data_0_split_1 I0403 19:05:33.109920 257810 net.cpp:408] data_data_0_split -> data_data_0_split_2 I0403 19:05:33.109930 257810 net.cpp:408] data_data_0_split -> data_data_0_split_3 I0403 19:05:33.109938 257810 net.cpp:408] data_data_0_split -> data_data_0_split_4 I0403 19:05:33.109949 257810 net.cpp:408] data_data_0_split -> data_data_0_split_5 I0403 19:05:33.109958 257810 net.cpp:408] data_data_0_split -> data_data_0_split_6 I0403 19:05:33.110057 257810 net.cpp:150] Setting up data_data_0_split I0403 19:05:33.110064 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110071 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110100 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110106 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110112 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110119 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110126 257810 net.cpp:157] Top shape: 24 3 300 300 (6480000) I0403 19:05:33.110132 257810 net.cpp:165] Memory required for data: 207360032 I0403 19:05:33.110138 257810 layer_factory.hpp:77] Creating layer conv0 I0403 19:05:33.110158 257810 net.cpp:100] Creating Layer conv0 I0403 19:05:33.110163 257810 net.cpp:434] conv0 <- data_data_0_split_0 I0403 19:05:33.110172 257810 net.cpp:408] conv0 -> conv0 terminate called after throwing an instance of 'cv::Exception' what(): OpenCV(3.4.9) /home/linqunshu/opencv/sourcecode/opencv-3.4.9/modules/imgproc/src/color.simd_helpers.hpp:88: error:(-2:Unspecified error) in function 'cv::impl::{anonymous}::CvtHelper<VScn, VDcn, VDepth, sizePolicy>::CvtHelper(cv::InputArray, cv::OutputArray, int) [with VScn = cv::impl::{anonymous}::Set<3, 4>; VDcn = cv::impl::{anonymous}::Set<3>; VDepth = cv::impl::{anonymous}::Set<0, 5>; cv::impl::{anonymous}::SizePolicy sizePolicy = (cv::impl::::SizePolicy)2u; cv::InputArray = const cv::_InputArray&; cv::OutputArray = const cv::_OutputArray&]'

Invalid number of channels in input image: 'VScn::contains(scn)' where 'scn' is 1

Aborted at 1585911933 (unix time) try "date -d @1585911933" if you are using GNU date PC: @ 0x7f59bfe52428 gsignal SIGABRT (@0x3e00003ef12) received by PID 257810 (TID 0x7f597d513700) from PID 257810; stack trace: @ 0x7f59bfe524b0 (unknown) @ 0x7f59bfe52428 gsignal @ 0x7f59bfe5402a abort @ 0x7f59c048c84d __gnu_cxx::verbose_terminate_handler() @ 0x7f59c048a6b6 (unknown) @ 0x7f59c048a701 std::terminate() @ 0x7f59c048a919 cxa_throw @ 0x7f59b56c9a2a cv::error() @ 0x7f59b56cba94 cv::error() @ 0x7f59b55878fa cv::errorNoReturn() @ 0x7f59b5588493 cv::detail::check_failedauto<>() @ 0x7f59b55896f2 cv::detail::check_failed_auto() @ 0x7f59b3c08833 cv::cvtColorBGR2HSV() @ 0x7f59b3a74977 cv::cvtColor() @ 0x7f59c1783e4c caffe::AdjustHue() @ 0x7f59c178868b caffe::RandomHue() @ 0x7f59c1789324 caffe::ApplyDistort() @ 0x7f59c17a7232 caffe::DataTransformer<>::DistortImage() @ 0x7f59c164d576 caffe::AnnotatedDataLayer<>::load_batch() @ 0x7f59c16133df caffe::BasePrefetchingDataLayer<>::InternalThreadEntry() @ 0x7f59c18039d5 caffe::InternalThread::entry() @ 0x7f59b2a435d5 (unknown) @ 0x7f599ac4a6ba start_thread @ 0x7f59bff2441d clone @ 0x0 (unknown) train.sh: line 10: 257810 Aborted (core dumped) ../../build/tools/caffe train -solver="solver_train.prototxt"-weights="mobilenet_iter_73000.caffemodel" -gpu 0`

PointCloudNiphon commented 4 years ago

what happened?

PointCloudNiphon commented 4 years ago

@chuanqi305 please help me, it confuse me for almost all the day, I can't find any useful solution in google,,

imistyrain commented 4 years ago

训练图中含有通道数为1的灰度图,检查下原图吧

PointCloudNiphon commented 4 years ago

谢谢你,我的数据集中是从coco里面提取的,里面会有灰度图吗?

PointCloudNiphon commented 4 years ago

训练图中含有通道数为1的灰度图,检查下原图吧 我的数据集中是从coco里面提取的,里面应该不会有灰度图吧