Aske-Rosted
Aske-Rosted
**Describe the bug** A clear and concise description of what the bug is. **To Reproduce** Run a script that triggers the `get_all_grapnet_classes` class. for an example running `graphnet/examples/04_training/03_train_dynedge_from_config.py` triggers the...
Adds 2 options for the weight fitter - a choice to employ an automatic log10 binning - a choice to cap the max single event weight to be x *...
Adds a combination of the RMSE loss and the von Mises-Fisher 3d loss, with the Mises-Fisher weighted with a factor 0.05. Similar to what was used in the 2nd place...
Adds a simple graph definition which simply does not construct a edge definition. Useful when the connection between the nodes is either not needed or constructed later in the model...
In regards to the PR #673 we should maybe consider adding a test that checks the what happens when data with empty events is fed to the graph builder, or...
**Is your feature request related to a problem? Please describe.** When working with high energy data the number of node activations becomes unreasonably large. Introducing a coarsening of the nodes...
the following is an implementation of the feature mentioned in #720. I had to include a type ignore on line 68. which I personally was not too happy with, but...
**Bug** The bug happens if less than 6 features is used in the FourierEncoder class. The script will _terminate_ with the following error > RuntimeError: It looks like your LightningModule...
**Automatic ensembling of databases when loading from config** For some use cases having several separate databases rather than one large db is preferable. However our current load from config only...
**Is your feature request related to a problem? Please describe.** There has been an increasing need for a "NodesAsDOMs" where nodes are constructed from summarizations of the DOMs. Specifically in...