3_ede_clustering_user_y3.yaml

Connector:
#  PREndpoint: 194.102.62.155
  Dask:
    SchedulerEndpoint: local
    Scale: 3
    SchedulerPort: 8787
    EnforceCheck: False
  MPort: 9200 # Moitoring port
  KafkaEndpoint: 10.9.8.136
  KafkaPort: 9092
  KafkaTopic: edetopic
#  Query: { "query": 'node_disk_written_bytes_total[5m]'}
  Query: {"query": '{__name__=~"node.+"}[1m]'}
  MetricsInterval: "1m" # Metrics datapoint interval definition
  QSize: 0
  Index: time
  QDelay: 10s # Polling period for metrics fetching
  Local: /Users/Gabriel/Dropbox/Research/ASPIDE/Datasets/ECI Chaos/Distributed Phase 1/finalized/single_node/training/df_anomaly.csv # Define the path to the local file for training

Mode:
  Training: True
  Validate: False
  Detect: False

Filter:
  DColumns:  # Which columns to delete
    - target
  Fillna: True # fill none values with 0
  Dropna: True # delete columns woth none values

Augmentation:
  Scaler: # if not used set to false
    StandardScaler:   # All scalers from scikitlearn
      copy: True
#      with_mean: True
#      with_std: True

# User defined clustering custom

#Training:
#  Type: clustering
#  Method: isoforest
#  Export: clustering_1
#  MethodSettings:
#    n_estimators: 10
#    max_samples: 10
#    contamination: 0.1
#    verbose: True
#    bootstrap: True

#Training:
#  Type: clustering
#  Method: !!python/object/apply:edeuser.user_methods.user_iso
#    kwds:
#      n_estimators: 100
#      contamination: 0.1
#      max_features: 10
#      n_jobs: -1
#      warm_start: False
#      random_state: 45
#      bootstrap: True
#      verbose: True
#      max_samples: 100
#  Export: clustering_y2

Training:
  Type: clustering
  Method: !!python/object/apply:edescikit.edepyod.ede_iso
    kwds:
      contamination: 0.01
      n_estimators: 100
      max_features: 1.0
      bootstrap: True
      n_jobs: -1
      verbose: 1
  Export: cluster_y2_v3

#Training:
#  Type: clustering
#  Method: !!python/object/apply:edescikit.edepyod.VAE_EDE
#    kwd:
#      contaminaton: 0.1
#      verbose: 1
#  Export: cluster_vae_y2

Detect:
  Method: IsolationForest
  Type: clustering
  Load: clustering_2
  Scaler: StandardScaler  # Same as for training



Misc:
  heap: 512m
  checkpoint: True
  delay: 10s
  interval: 30m
  resetindex: False
  point: False