SSD-on-Custom-Dataset
SSD-on-Custom-Dataset copied to clipboard
Coldmooon
Error during evaluation phase
Issue summary
Cannot copy param 0 weights from layer 'conv4_3_norm_mbox_loc'; shape mismatch. Source param shape is 12 512 3 3 (55296); target param shape is 16 512 3 3 (73728). To learn this layer's parameters from scratch rather than copying from a saved net, rename the layer. *** Check failure stack trace: *** @ 0x7f8cf64325cd google::LogMessage::Fail() @ 0x7f8cf6434433 google::LogMessage::SendToLog() @ 0x7f8cf643215b google::LogMessage::Flush() @ 0x7f8cf6434e1e google::LogMessageFatal::~LogMessageFatal() @ 0x7f8cf6cf73cb caffe::Net<>::CopyTrainedLayersFrom() @ 0x7f8cf6cfdc25 caffe::Net<>::CopyTrainedLayersFromBinaryProto() @ 0x7f8cf6cfdcbe caffe::Net<>::CopyTrainedLayersFrom() @ 0x40a519 CopyLayers() @ 0x40b974 train() @ 0x407590 main @ 0x7f8cf53a2830 __libc_start_main @ 0x407db9 _start @ (nil) (unknown) Aborted (core dumped)
.
Your system configuration
Operating system: Ubuntu 16.04 Compiler: CUDA version (if applicable): Cuda 8.0 CUDNN version (if applicable): CUDNN 5.1 BLAS: Atlas Python or MATLAB version (for pycaffe and matcaffe respectively): Python 2.7
It has already been clear. It's a shape mismatch problem. Modify your prototxt file if you want to finetune conv4, or rename the layer if you want to learn this layer's parameters from scratch.
Hello actually I already renamed the layer but right now I'm wondering why I cant find the comp4_det_test file does it only generate after i finished all iterations? I'm wondering what I did wrong, does it only generate after finishing all iterations? Will appreciate your help Thank you very much
@seandales This is a different question. Yes, whenever the model is tested, some comp4_det_test files are outputted. These files are in your data directory which is defined by the output_result_dir keyword in examples/ssd/ssd_pascal.py.