-
STEP 1 Calculate regression coefficient. The regression coefficient indicates how well linear regression fits to the original data.
-
STEP 2 The program removes some data numbers if the coefficient regression get worst, with these numbers.
Then, you get "clean data", without points that generate the bad coefficient.
Example:
Pairs (1,1) (2,2) ( 3,3 ) (4,7) (5,5) (6,6)
Should delete (4,7)
Design Patterns:
- Command , undo Command
SOLID principles: - Dependency Injection
Continous Integration: - Travis, Coveralls
Testing: - Valgrind: memory leaks - Unit testing with google test
Technology: - Docker, c++