Daniel C. Jones

The branch on github was a bit out of date. If you pull the latest changes you should able to build it now. I plan to eventually merge this into...

Thanks! I'm going to keep this issue open for now, so others that run into this problem have some context.

You should be able to something like this to install the experimental branch ```sh git clone https://github.com/dcjones/proseg.git cd proseg git checkout poisfa cargo install --path . ``` It's is very...

Seems like others are having this issue: https://github.com/rust-ml/linfa/issues/307 I'll see if I can figure out a work around on my end.

Hi @alikhuseynov, sure thing. This branch tries to overcome the the issues the current version of Proseg has in high dimensionality settings by implementing probabilistic matrix factorization as part of...

I've tried pinning sprs at 0.11.1 in the `poisfa` branch, which will hopefully fix the compilation errors. I'll see if there's an easy way to avoid having a gfortran dependency...

Right now proseg can only be informed by a fixed reassignment penalty from prior transcript assignments. The two options to try tweaking are: `--prior_seg_reassignment_prob` and `--nuclear_reassignment_prob`, where lowering these more...

Excellent! Let me know if I can do anything to make that easier.

As a heads up for those trying to run higher-plex data, I pushed a new branch `proseg2` which has further work on this model and is rebased on main. I'll...

Yep, this new version does include the changes in the `proseg2` branch. I would recommend using it with very any dataset now, with generally no need to tinker with settings....