Skip to content

Added PFN models and reorganized the repository structure #8

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cmsbuild merged 1 commit into from
Mar 14, 2025
Merged

Added PFN models and reorganized the repository structure #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,10 +25,17 @@
- `superclustering/regression_v1.onnx`: DNN for supercluster energy regression. Input format : `batch x 8 (features)`. Output : `batch x 1` (supercluster regressed energy). Used in `RecoHGCal/TICL/plugins/EGammaSuperclusterProducer.cc`.
- `ticlv5/onnx_models/`: The models are trained based on TICLv5 reconstruction information using a simple CNN-based approach. Two models have been trained separately: one for trackster energy regression and one for particle ID. These models are saved in ONNX format for time optimization.
- `Common input tensor`: Both models share the same initial input tensor, dimensions batch x 50 (layers) x 10 (clusters) x 3 (features).
- `ticlv5/patternrecognition/id_v*.onnx`:
- `ticlv5/DNN/patternrecognition/id_v*.onnx`:
- `"input"`: Input tensor with dimensions batch x 50 (layers) x 10 (clusters) x 3 (features).
- `"output/pid_output"`: Output tensor with dimensions batch x 8 representing particle ID probabilities (from a softmax output). The probabilities refer to: photon, electron, muon, neutral pion, charged hadron(pion), neutral hadron(kaon), ambiguous, and unknown cases (in that order). The probabilities help in classifying the particle based on its type, distinguishing between hadronic and electromagnetic categories.
- `ticlv5/linking/energy_v*.onnx`:
- `ticlv5/DNN/linking/energy_v*.onnx`:
- `"input"`: Input tensor with dimensions batch x 50 (layers) x 10 (clusters) x 3 (features), concatenated with the output of the particle ID model ("output/pid_output").
- `"output/enreg_output"`: Output Tensor with dimension batch x 1 (regressed energy). This value represents the trackster energy as estimated by the model based on the training data, compared to the true and reconstructed energies of the particle.

- `Common input tensor`: Both models share the same initial input tensor, dimensions batch x 50 (layers) x 10 (clusters) x 7 (features).
- The PFN models are designed to be trained on both low-level variables, such as layer cluster features, and high-level variables, such as trackster features.
- `ticlv5/PFN/patternrecognition/id_v*.onnx`:
- `"input"`: Input tensor with dimensions batch x 50 (layers) x 10 (clusters) x 7 (features).
- `"output/pid_output"`: Output tensor with dimensions batch x 8 representing particle ID probabilities (from a softmax output). The probabilities refer to: photon, electron, muon, neutral pion, charged hadron(pion), neutral hadron(kaon), ambiguous, and unknown cases (in that order). The probabilities help in classifying the particle based on its type, distinguishing between hadronic and electromagnetic categories.
- `ticlv5/PFN/linking/energy_v*.onnx`:
- `"input"`: Input tensor with dimensions batch x 50 (layers) x 10 (clusters) x 7 (features), concatenated with the output of the particle ID model ("output/pid_output").
- `"output/enreg_output"`: Output Tensor with dimension batch x 1 (regressed energy). This value represents the trackster energy as estimated by the model based on the training data, compared to the true and reconstructed energies of the particle.
0 ticlv5/onnx_models/linking/energy_v0.onnx → ...v5/onnx_models/DNN/linking/energy_v0.onnx
View file
Open in desktop
File renamed without changes.
0 ticlv5/onnx_models/linking/id_v0.onnx → ticlv5/onnx_models/DNN/linking/id_v0.onnx
View file
Open in desktop
File renamed without changes.
0 ..._models/patternrecognition/energy_v0.onnx → ...els/DNN/patternrecognition/energy_v0.onnx
View file
Open in desktop
File renamed without changes.
0 ...onnx_models/patternrecognition/id_v0.onnx → ..._models/DNN/patternrecognition/id_v0.onnx
View file
Open in desktop
File renamed without changes.
Binary file added ticlv5/onnx_models/PFN/linking/energy_v0.onnx
View file
Open in desktop
Binary file not shown.
Binary file added ticlv5/onnx_models/PFN/linking/id_v0.onnx
View file
Open in desktop
Binary file not shown.
Binary file added ticlv5/onnx_models/PFN/patternrecognition/energy_v0.onnx
View file
Open in desktop
Binary file not shown.
Binary file added ticlv5/onnx_models/PFN/patternrecognition/id_v0.onnx
View file
Open in desktop
Binary file not shown.