You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The MLPF algorithm has not been officially introduced in CMSSW yet. Implementation in a SonicTriton producer is likely the easiest path (especially for a PyTorch graph-based network, since the PyTorch C++ API is not yet available in CMSSW).
Particular considerations:
Consider introducing ProcessModifier to enable the alternative sequence
The initial proof of concept MLPF producer was merged in cms-sw/cmssw#32048.
I'm working on a MLPF v2 training and PR for cmssw with the goal of reasonable downstream physics with the MLPF candidates in CMSSW. I think additional inference implementations i.e. Sonic could be a good technical follow-up to that. This may be a month or so from now.
The MLPF algorithm has not been officially introduced in CMSSW yet. Implementation in a SonicTriton producer is likely the easiest path (especially for a PyTorch graph-based network, since the PyTorch C++ API is not yet available in CMSSW).
Particular considerations:
Assigned to: @jpata
The text was updated successfully, but these errors were encountered: