Andrew McNutt

In our most recent paper, [GNINA 1.0](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-021-00522-2), we evaluated the average time for docking a single protein-ligand pair with `cnn_scoring` set to `refinement` and `rescore` in comparison to the default...

Our benchmarks were ran on the PDBbind Core Set v. 2016 utilizing a workstation with a dual 16-core 2.3 GHz Intel Xeon 5218 with 96 GB of RAM and a...

The weights files for all default2018 models have been uploaded with [(gnina/models#31)](https://github.com/gnina/models/pull/31). Training of `crossdock_default2018` requires the [CrossDock2020 dataset](https://github.com/gnina/models/tree/master/data/CrossDocked2020). `redock_default2018` is trained on the redocked subset of the same dataset.

`GetBestRMS` calculates RMSD **and** then aligns them in space. `CalcRMS` calculates the RMSD without moving either molecule.

> Dear professor and developers, I come up against a few questions about the procedures in training. > > 1. When I split my whole new dataset in train&validation, I...

> I want to know more about the output `rmsd`, I didn't find an explanation for the `default2018.***.rmsd.test` number's meaning in the README. > >  I am not sure...