Repository files navigation

PyTorch implements "DeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning"

https://colab.research.google.com/drive/1xWHovxTkysiPei68bTywqB2oCo6JelAU?usp=sharing

1. Build environment

   git clone https://github.com/nailo2c/deeplog.gitcd deeplogpython3 -m venv venv. venv/bin/activatepip install -r requirements.txt

2. Run local example

   We use open dataOpenStack from logpai's loghub

   2.1. Preprocess

   cdexample/python3preprocess.py

   2.2. Train

   num-class is count ofevent_id_map, whereevent_id_map is generated bypreprocess.py.num-candidates is self-define, here we definenum-candidates isnum-class*0.1

   python3train.py--num-class1143--num-candidates114--epochs35--window-size3--localTrue

   2.3. Predict

   python3predict.py--threshold25

3. Result

   Accuracy	0.9430014
   Precision	0.6497461
   Recall	0.9275362
   F1	0.7641791

4. Deactivate

   deactivate

• This tree structure is generate by mac terminal tooltree & copy paste it toREADME.md.

  tree -I"__pycache__|tmp.*">> tmp.txt

.├── README.md├── deeplog│   ├── __init__.py│   └── deeplog.py├── example│   ├── data│   │   └── OpenStack│   │       ├── anomaly_labels.txt│   │       ├── openstack_abnormal.log│   │       ├── openstack_normal1.log│   │       └── openstack_normal2.log│   ├── predict.py│   ├── preprocess.py│   └── train.py└── requirements.txt