Closed theodore3131 closed 6 years ago
Two solutions:
v 1 0 dog->v 1 0 1
v 1 1 cat->v 1 1 2
And of course you need another script to transform ids to strings on the result.
database.cc
, and change the format of label which resides on the last column. If you use C++ "string", I expect that you just need to change some lines and get it work very quickly.Thank you so much for your solution, I will try it out.
Hi, recently I come across a problem again. For a input graph file which is bigger than 100MB, the gbolt would crash because of segmentation fault (core dumped), can you give me any suggestions about this? Thanks again!
Hi, if it is a core dump, you should send me the core dump file and your binary, or I couldn't figure out the causes.
You can also locate the core dump position and post it on our discussion thread.
It reads like follows on the terminal:
I0506 12:52:51.015228 3380 gbolt.cc:37] gbolt read input time: 1.61415
I0506 12:52:53.075754 3380 gbolt_execute.cc:27] gbolt construct graph time: 2.06035
I0506 12:52:53.077332 3380 gbolt_execute.cc:54] gbolt thread 0 create
Segmentation fault (core dumped)
I think you misunderstand my instructions.
Please send me your data and your arguments to run the program. I will try to figure it out.
Oh, sorry. Here are my commands and the file:
export OMP_NUM_THREADS=1
./build/gbolt -input_file extern/data/rgraph -support 0.05 -output_file ./result0 -pattern=true
The bug is caused by your rgraph data.
In rgraph, you have total the number of graphs different from the last graph id given by indicator "t #". Please examine your data and make sure there's no "hole" between two graph ids.
I will also modify the code to make it more robust.
Problem fixed.
Also please delete the last "t # -1" entry from rgraph.
This do works and saves me a lot of time. Thank you so much for your kind help!
You're welcome, please let me know if you have further questions.
Hello, for the database that I need to deal with in my research, the label of edges are strings instead of integers, could you give me some instructions about how to modify the source code so that I can use this for my database? Thanks!