Dean
/
mlops-best-practices
mirror of https://github.com/karan842/mlops-best-practices.git


  
1

	
2

	
3

	
4

	
5

	
6

	
7

	
8

	
9

	
10

	
11

	
12

	
13

	
14

	
15

	
16

	
17

	
18

	
19

	
20

	
21

	
22

	
23

	
24

	
25

	
26

	
27

	
28

	
29

	
30

	
31

	
32

	
33

	
34

	
35

	
36

	
37

	
38

	
39

	
40

	
41

	
42

	
            base:
  project: mlops-best-practices
  author: Karan Shingde
  email: karanshingde@gmail.com
  version: 0.0.1

data_source:
  type: csv
  raw_data_path: raw_data/data.csv
  generated_data_path: synthetic_data/synthetic_data.csv
  train_data_path: artifacts/train.csv
  test_data_path: artifacts/test.csv

data_transformation:
  target_column: "Exited"

  numerical:
    columns: ['CreditScore', 'Age', 'Tenure', 'Balance', 'NumOfProducts', 'HasCrCard', 'IsActiveMember', 'EstimatedSalary']
    # scaler: StandardScaler(with_mean=False)
    # imputer: SimpleImputer(strategy='mean')
    
  categorical: 
    columns: ['Geography', 'Gender']
    # transformer: OneHotEncoder()
    # imputer: SimpleImputer(strategy='most_frequent')
    # encoder: OneHotEncoder()
    # scaler: StandardScaler(with_mean=False)

  imbalance_threshold: 60

  outliers:
    quantile1: 0.25
    quantile2: 0.75

  test_split: 0.2
  random_state: 42

artifacts:
  model_path: artifacts/model.pkl
  preprocessor_path: artifacts/preprocessor.pkl
  params_path: artifacts/params.yaml