-
IMPORTING THE DATASETS:
Imported the training and testing datasets using pandas .read_csv() function. -
EXPLORATORY DATA ANALYSIS:
a)Placed all the missing values in a seperate list.
b)Placed all thenumeric, categorical and discrete values in different lists.
c)Visualized numerical features with line plot from matplotlib.
d)Visualized continousfeatures with histogram from matplotlib. -
FEATURE ENGINEERING:
a) Replace all the missing values with the mean of their respective field.
b) Encoded Object datatypes using Label Encoder from sklearn -
FEATURE SCALING:
a) Scaled all the data with standard scaler from sklearn. -
MODEL FITTING:
a) Split the data into training and test sets.
b)Trained the model using XGBoost Regressor.
c) hyperparameter tuning with RandomsearchCV.
d) Stored the predictions in 'MySubmission.csv'.