You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

build_pipelines.sh 671 B

1234567891011121314151617181920212223242526
  1. #!/bin/bash
  2. if [ ! -d "./pipelines/simple" ]; then
  3. mkdir -p ./pipelines/simple
  4. fi
  5. if [ ! -d "./pipelines/subseq" ]; then
  6. mkdir -p ./pipelines/subseq
  7. fi
  8. python pipeline_construction/pipeline_construction_simple.py
  9. python pipeline_construction/pipeline_construction_subseq.py
  10. data="swan_sf creditcard web_attack water_quality"
  11. for d in $data
  12. do
  13. if [ ! -d "./pipelines/AE/$d" ]; then
  14. mkdir -p ./pipelines/AE/$d
  15. fi
  16. if [ ! -d "./pipelines/RNN_LSTM/$d" ]; then
  17. mkdir -p ./pipelines/RNN_LSTM/$d
  18. fi
  19. python pipeline_construction/neural/build_AE_pipeline.py "./data/"$d".csv"
  20. python pipeline_construction/neural/build_RNNLSTM_pipeline.py "./data/"$d".csv"
  21. done

全栈的自动化机器学习系统,主要针对多变量时间序列数据的异常检测。TODS提供了详尽的用于构建基于机器学习的异常检测系统的模块,它们包括:数据处理(data processing),时间序列处理( time series processing),特征分析(feature analysis),检测算法(detection algorithms),和强化模块( reinforcement module)。这些模块所提供的功能包括常见的数据预处理、时间序列数据的平滑或变换,从时域或频域中抽取特征、多种多样的检测算