darknet2caffe.py sets convolution_param['pad'] = str(int(convolution_param['kernel_size'])/2), but if the kernel size is not cleanly divisible by 2, this results in an error
[libprotobuf ERROR C:\Users\guillaume\work\caffe-builder\build_v140_x64\packages\protobuf\protobuf_download-prefix\src\protobuf_download\src\google\protobuf\text_format.cc:298] Error parsing text-format caffe.NetParameter: 14:14: Expected integer, got: 1.5
WARNING: Logging before InitGoogleLogging() is written to STDERR
F0403 23:57:00.668103 11120 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: data\yolo-obj.prototxt
There are a few other instances (e.g. you may get decimals during "step 3" where reorg layer calculations occur). I wrapped them in int() but it's unclear if this is Right(tm).
darknet2caffe.py sets
convolution_param['pad'] = str(int(convolution_param['kernel_size'])/2), but if the kernel size is not cleanly divisible by 2, this results in an errorThere are a few other instances (e.g. you may get decimals during "step 3" where reorg layer calculations occur). I wrapped them in
int()but it's unclear if this is Right(tm).Sample .cfg https://gist.github.com/riverar/46872ab59c4cbd3c56dda4fe61b4f9e3