update readme & fix pep8 problem for WSSAlgorithm

README.md

@@ -1,7 +1,33 @@
-# Machine-Learnings-Methods-for-Time-Series-Data-Analysis at Oil And Gas Industry.
+# Time-Series Change Point Detection models for Oil and Gas Industries.
 
-# REFACTORING HERE!!
-## will be available soon
+#### version: 0.0.1
+- Syth data generator as well as real data available at data dir.
+- Refactored models based on KalmanFilter and Singular Value Decomposition technique avalable at model dir.
+- Custom fast optimization based on heuristics methods.
+- Streamlit example app at examples dir.
+
+#### Notes:
+ - Most of the implemented idea/code based on my master thesis. SingularSequenceTransformation and WindowSizeSearch 
+optimization 
+classes has been refactored based on implementations from ***Fedot.Industrial*** legacy.
+ - To score change point detection models functions from ***TSAD*** lib has been used\adopted.
+
+#### Tasks:
+
+## To set up local project dependencies:
+```commandline
+ python3 setup.py build
+```
+
+TO DO:
+
+1. Docker images for example files at examples dir.
+2. Hybrid model
+3. Evolution optimization for model hyperparameters.
+4. default notebook examples
+5. FastAPI service for end-to-end use in container.
+6. Advanced change point detection models.
+7. More tests.
 
 [//]: # (Here you find notebooks with Change Point Detection methods in Petroleum Data. Mainly I focus to experiment with Fedot.Industrial library.)
 

optimization/WSSAlgorithms.py

@@ -16,7 +16,7 @@ class WindowSizeSelection:
  wss_algorithm: str
  type of WSS algorithm. It can be 'highest_autocorrelation', 'dominant_fourier_frequency',
  'summary_statistics_subsequence' or 'multi_window_finder'.
- By default it is 'dominant_fourier_frequency'
+ By default, it is 'dominant_fourier_frequency'
  Reference:
  (c) "Windows Size Selection in Unsupervised Time Series Analytics: A Review and Benchmark. Arik Ermshaus,
  Patrick Schafer, and Ulf Leser"