Merge branch 'main' of hf.co:jpata/particleflow into main

clic/clusters/v1.9.0/README.md

@@ -4,6 +4,26 @@ This model reconstructs particles in a detector, based on the tracks and calorim
 
 ## Model Details
 
+The performance is measured with respect to generator-level jets and MET computed from Pythia particles, i.e. the truth-level jets and MET.
+
+<details>
+  <summary>Jet performance</summary>
+  
+  <img src="plots_checkpoint-26-2.004527/clic_edm_ttbar_pf/jet_response_iqr_over_med_pt.png" alt="ttbar jet resolution" width="300"/>
+  <img src="plots_checkpoint-26-2.004527/clic_edm_qq_pf/jet_response_iqr_over_med_pt.png" alt="qq jet resolution" width="300"/>
+  <img src="plots_checkpoint-26-2.004527/clic_edm_ww_fullhad_pf/jet_response_iqr_over_med_pt.png" alt="ttbar jet resolution" width="300"/>
+
+</details>
+
+<details>
+  <summary>MET performance</summary>
+  
+  <img src="plots_checkpoint-26-2.004527/clic_edm_ttbar_pf/met_response_iqr_over_med.png" alt="ttbar MET resolution" width="300"/>
+  <img src="plots_checkpoint-26-2.004527/clic_edm_qq_pf/met_response_iqr_over_med.png" alt="qq MET resolution" width="300"/>
+  <img src="plots_checkpoint-26-2.004527/clic_edm_ww_fullhad_pf/met_response_iqr_over_med.png" alt="ttbar MET resolution" width="300"/>
+
+</details>
+
 ### Model Description
 
 - **Developed by:** Joosep Pata, Eric Wulff, Farouk Mokhtar, Mengke Zhang, David Southwick, Maria Girone, David Southwick, Javier Duarte, Michael Kagan
@@ -54,10 +74,11 @@ The following datasets were used:
 /eos/user/j/jpata/mlpf/tensorflow_datasets/clic/clic_edm_ww_fullhad_pf/2.2.0
 ```
 
-The truth and target definition was updated in [jpata/particleflow#345](https://github.com/jpata/particleflow/pull/345) have an updated truth and target definition with respect to [Pata, J., Wulff, E., Mokhtar, F. et al. Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors. Commun Phys 7, 124 (2024)](https://doi.org/10.1038/s42005-024-01599-5).
+The truth and target definition was updated in [jpata/particleflow#345](https://github.com/jpata/particleflow/pull/345) with respect to [Pata, J., Wulff, E., Mokhtar, F. et al. Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors. Commun Phys 7, 124 (2024)](https://doi.org/10.1038/s42005-024-01599-5).
 
-In particular, target particles for MLPF reconstruction are based on status=1 particles.
-For non-interacting status=1, nearby (dR<0.2) interacting status=0 are used instead.
+In particular, target particles for MLPF reconstruction are based on `status=1` particles.
+For non-interacting `status=1`, nearby (dR<0.2) interacting `status=0` are used instead.
+It's important to note that truth and target jets are defined in the center of mass frame, whereas PF particles are defined in the lab frame: https://github.com/key4hep/k4geo/issues/399#issuecomment-2381714391.
 
 The datasets were generated using Key4HEP with the following scripts:
 - https://github.com/HEP-KBFI/key4hep-sim/releases/tag/v1.0.0
@@ -65,6 +86,9 @@ The datasets were generated using Key4HEP with the following scripts:
 
 ## Training Procedure 
 
+<details>
+  <summary>Training script</summary>
+  
 ```bash
 #!/bin/bash
 #SBATCH --job-name=mlpf-train
@@ -111,7 +135,13 @@ singularity exec \
      --train --gpu-batch-multiplier 128 --num-workers 8 --prefetch-factor 100 --checkpoint-freq 1 --conv-type attention --dtype bfloat16 --lr 0.0001 --num-epochs 30
 ```
 
+</details>
+
 ## Evaluation
+
+<details>
+  <summary>Evaluation script</summary>
+  
 ```bash
 #!/bin/bash
 #SBATCH --partition gpu
@@ -128,10 +158,11 @@ singularity exec -B /scratch/persistent --nv \
      --env KERAS_BACKEND=torch \
      $IMG  python3 mlpf/pyg_pipeline.py --dataset clic --gpus 1 \
      --data-dir /scratch/persistent/joosep/tensorflow_datasets --config parameters/pytorch/pyg-clic.yaml \
-     --test --make-plots --gpu-batch-multiplier 100 --load $WEIGHTS --dtype bfloat16 --prefetch-factor 10 --num-workers 8 --load $WEIGHTS --ntest 50000
-~               
+     --test --make-plots --gpu-batch-multiplier 100 --load $WEIGHTS --dtype bfloat16 --prefetch-factor 10 --num-workers 8 --load $WEIGHTS --ntest 50000            
 ```
 
+</details>
+
 ## Citation
 
 ## Glossary
@@ -142,4 +173,4 @@ singularity exec -B /scratch/persistent --nv \
 
 ## Model Card Contact
 
-Joosep Pata, joosep.pata@cern.ch
+Joosep Pata, joosep.pata@cern.ch