Open lila167 opened 3 years ago
Thank you for the feedback @lila167.
@adyprat is there a way to control the noise from a command-line parameter? I couldn't see one after a quick glance.
The noise is currently controlled by a hard-coded value in this function https://github.com/Murali-group/BoolODE/blob/b4a752a5ca5087f65d6a58802785b4590c47e29e/BoolODE/simulator.py#L3 @lila167 we chose to hard-code a nominal value for the strength of noise because larger values did not always yield the expected number of steady states. We have not systematically tried varying the noise parameters to lower values.
Thanks for explaining, @amoghpj. Is it worthwhile to make it a command line parameter and add your warnings to the help message that goes with the parameter definition?
@lila167 you can of course change the hardcoded value to whatever you desire.
Thank you both!! I will give it a try with lower noise level.
Thanks for explaining, @amoghpj. Is it worthwhile to make it a command line parameter and add your warnings to the help message that goes with the parameter definition?
Yes, I think it is a good idea to expose this parameter. I will create a pull request with the additions.
@amoghpj Checking to see if you had time to create a PR for this feature.
Will work on this this week. Thanks for the reminder!
I've created a pull request here https://github.com/Murali-group/BoolODE/pull/17/commits/e8835f9bb104f3bf539ab467ad8703371a67a920 . It would be great if someone can test the patch and confirm that the expected behavior.
Hi,
Thank you very much for this tool!! I am wondering if one can generate the curated datasets (Fig. 3 of your paper) with less noise.
Best