Missing Entry Script, Parameter Queries, and Performance Alignment

[CGaaaaaa]

I hope this message finds you well. I’ve been working with StarBASE-GP and have encountered a couple of issues I’m hoping to seek your guidance on.
First, I noticed that the entry code script seems to be missing from the resources provided. Additionally, I’m a bit unclear about the specific roles of "cores" and "bin_size" in the framework—could you kindly elaborate on how these parameters function and how they might impact performance?
I’ve been running the tool following the documentation, but unfortunately, the results I’m getting aren’t matching the performance outlined in your paper. I suspect I might be overlooking something in the setup or parameter configuration.
Any insights or assistance you could provide to help resolve these issues would be deeply appreciated. Thank you so much for your time and support!

[jgh9094]

We are working on pushing updates to the 'dev' branch soon with all the latest changes. What's on the current dev should work but we are still refining things. You are correct that there is still some variables from prior implementations on the main branch.

'bin_size' is no longer part of StarBASE, we instead use a sliding window for more fine-grain resolution. The sliding window could be improved with more id tracking, but is not much of a slowdown.

'cores' specifies the number of computations that can be done in parallel. We use ray to parallelize much of the regression and encoding. As such, the more cores you give, the faster computations can be done.

The results from our preprint will not align with the current implementation.