What are the meanings of meta.write_number and meta.total_write_number?

[walkccc]

Hello, thanks for this great model first.

I have a few questions.
Is it possible to train the data with my personal data?
I've trained it with my personal data, but the meta.write_number is equal to 0 all the time.
I want to know what's the difference between meta.write_number and meta.total_write_number?

Thank you very much!

This is part of my logs:

I0827 17:11:04.304325 6117 cpm_data_transformer.cpp:73] dataset: COCO; img_size: [427 x 640]; meta.annolist_index: 44; meta.write_number: 0; meta.total_write_number: 41; meta.epoch: 0
I0827 17:11:04.588376 6107 net.cpp:150] Setting up conv1_1
I0827 17:11:04.588404 6107 net.cpp:157] Top shape: 4 64 368 368 (34668544)
I0827 17:11:04.588409 6107 net.cpp:165] Memory required for data: 199411840
I0827 17:11:04.588433 6107 layer_factory.hpp:77] Creating layer relu1_1

[Hill-Gu]

@walkccc I want to train my personal data, when I install the caffe_train using "make all -j64", the "make: *** [.build_release/src/caffe/layers/cpm_data_layer.o] Error 1" error occured, what should I do to solve this problem?Looking forward your reply, thanks!