jetclass_classifier_epic.yaml

# @package _global_

# to execute this experiment run:
# python train.py experiment=jetclass_classifier

defaults:
  - override /data: classifier_data_jetclass.yaml
  # - override /model: mlp_classifier.yaml
  # - override /model: particlenet_classifier.yaml
  # - override /model: ParT_classifier.yaml
  - override /model: epic_classifier_jetclass.yaml
  - override /callbacks: jetclass_classifier.yaml
  - override /trainer: gpu.yaml

# all parameters below will be merged with parameters from default configurations set above
# this allows you to overwrite only specified parameters

# add here checkpoint to continue training
# ckpt_path: XXX/checkpoints/last-EMA.ckpt

tags: ["fm-classifier_test", "JetClass", "ClassifierTest"]
run_note: ""
seed: 122

vars:
  epochs: 200
  warmup: 3
  val_check_interval: null

data:
  # kin_only: true
  batch_size: 512
  # number_of_jets: 100000
  use_weaver_axes_convention: false
  data_file: XXX.h5
  pf_features_list:
    - part_etarel
    - part_dphi
    - log_part_pt
    - log_part_energy
    - log_part_ptrel
    - log_part_energyrel
    - part_deltaR
    - part_charge
    - part_isChargedHadron
    - part_isNeutralHadron
    - part_isPhoton
    - part_isElectron
    - part_isMuon
    - tanh_part_d0val
    - part_d0err
    - tanh_part_dzval
    - part_dzerr
  # debug_sim_only: true
  used_flavor: Tbqq

# setting load_weights_from will load the weights from the given checkpoint, but start training from scratch
# load_weights_from: XXX.ckpt

model:
  net_config:
    dropout: 0.3
    input_dim: 17
    out_dim: 2
    # num_points: 200 # top tagging landscape
#   input_dims: 7  # ParticleNet
#   input_dim: 7  # ParT
#   num_classes: 2
#   scheduler:
#     warmup: ${vars.warmup}
#     max_iters: ${vars.epochs}
#   optimizer:
#     lr: 0.0001

#early_stopping:
#  monitor: "val/loss"
#  patience: 2000
#  mode: "min"

task_name: "jetclass_classifier"

trainer:
  min_epochs: 1
  max_epochs: ${vars.epochs}
  val_check_interval: ${vars.val_check_interval}
  gradient_clip_val: 0.5

logger:
  wandb:
    tags: ${tags}
    group: "flow_matching_jetclass"
    name: ${task_name}
  comet:
    experiment_name: null
    project_name: "flow-matching-classifierTest"