ymirsky / VulChecker

A deep learning model for localizing bugs in C/C++ source code (USENIX'23)
GNU General Public License v3.0
115 stars 12 forks source link

Minimum system requirements #12

Open alcastronic opened 5 days ago

alcastronic commented 5 days ago

Hi @ymirsky thanks for providing this repository.

I downloaded the VM and imported it. Unfortunately, the "model_train_demo.sh" aborts. According to the message in the script, this is due to insufficient memory.
What is the minimal amount of memory needed for the "model_train_demo.sh" script to run?

I have configured the vulchecker VM with 86GB of memory. Below is the output of the script ran in the VM.

$ ./model_train_demo.sh

[truncated]

Note: if you get an 'abort' or 'killed' then the VM has run our of memory
Loading train from cache
Loading test

Aborted!

Are you aware of any other instances that provoke the "Abort!" message, then insufficient amount of memory?

Note: The other demo scripts "preprocess_demo.sh" and "model_prediction_demo.sh" run.

alcastronic commented 4 days ago

I have tested the training with a different dataset for TEST, which worked successfully. I'll keep this issue open for now, since the failed command was in the demo script and I could not reproduce it.