bi.clique not working as expected

[flin2]

Hi Yuping,

I also encountered the "memory not mapped" issue, and after I looked at the previous threads, I thought maybe my file was too large (it was about 800x60 but the job still failed even if I ran it with 500GB of memory and limited the range to search!). So I tried to run bi.clique with getclique set to 0, but even when I used the exact file and code found in this example, I still got an error saying

Error in data.raw[[1]][[i]] : subscript out of bounds Calls: bi.clique -> unlist

Do you have any idea what might have caused this? For your reference, I installed the package from CRAN, and ran the code both on Windows and a linux server. Thank you!

Best, Feng

[YupingLu]

Hi Feng,

Sorry about the late reply. I'm busy with dissertation recently. I doubt your file format is not correct. Can you send an example file you are using to me?

Thanks, Yuping

[flin2]

Hello Yuping,

Sorry for my delayed reply and hope everything has been well. I just tried to use the package again but I got this issue again.

caught segfault address 0x488816e4, cause 'memory not mapped' Traceback: 1: bi.clique(f_raw, 20, 50) An irrecoverable exception occurred. R is aborting now ...

I am attaching the data file that I used. I ran the program on a linux server with R 3.5.3 and biclique 1.0.2.

Best, Feng wbg_bigraph.zip

[YupingLu]

Hi Feng,

I'll try your data later today and get back to you soon.

Thanks, Yuping

On Wed, Jul 17, 2019 at 12:09 PM Feng notifications@github.com wrote:

Hello Yuping,

Sorry for my delayed reply and hope everything has been well. I just tried to use the package again but I got this issue again.

caught segfault address 0x488816e4, cause 'memory not mapped' Traceback: 1: bi.clique(f_raw, 20, 50) An irrecoverable exception occurred. R is aborting now ...

I am attaching the data file that I used. I ran the program on a linux server with R 3.5.3 and biclique 1.0.2.

Best, Feng wbg_bigraph.zip https://github.com/YupingLu/biclique/files/3402773/wbg_bigraph.zip

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/YupingLu/biclique/issues/8?email_source=notifications&email_token=AACEVSVCUSM6I4LVG7YEUNLP7472HA5CNFSM4HIJAXI2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD2E476A#issuecomment-512348152, or mute the thread https://github.com/notifications/unsubscribe-auth/AACEVSSFDYY4GN22RVD5H3TP7472HANCNFSM4HIJAXIQ .

[YupingLu]

Hi Feng,

I checked your file. R requires a lot of memory. You got this error because your computer was out of memory. You can check my previous answer here. https://github.com/YupingLu/biclique/issues/7#issuecomment-480601476

Thanks, Yuping

[flin2]

Hi Yuping,

Recently I am trying to use this package again, but I still ran into this memory issue. You mentioned in #7 that you also have an implementation in C++ that would allow the output of all bicliques, so I am wondering if you could direct me to that. Thank you!

Best, Feng

[YupingLu]

Hi Feng,

What's your email address? I can send the code to you via email.

Thanks, Yuping

[flin2]

Hi Yuping,

Thanks for your reply! Could you send it to fenglin.booth@gmail.com? In fact I have modified your code a little bit and managed to export one biclique at a time, but I think it would also be good to have the cpp code. In addition, I had a look at your source code and wondering if line 341 in biclique.c has a typo--I thought here you intended to call biclique_find_improve (instead of biclique_find_basic)?

Best, Feng

[dpuelz]

Hi @flin2! I am wondering if you are able to share your code that exports one Biclique at a time? Alternatively, is there an easy modification to do this in R? Thanks