Mobile Price Range Classifier

Steps that I performed

• Study,collect and import the data.
• Perform EDA(exploratory data engineering) on it to get better visualization
  • get info about data using df.describe()
  • checking na values by inbuilt function and custom one.
  • Find correlation between features and output.
  • Visualize it using seaborn pairplot or scatter plot.
  • Visualzing some features separately.
• Performing Feature engineering.
  • dropping columns which has less correlation and has not anything to do with our label
  • creating derived features
  • performing train_test_split
• Feature scaling
  • standard scaling
  • min max scaling
• using pipeline
• Selecting the models
  • knn classfier
  • decision tree classifier
  • random forest tree clasfier
• training and evaluating models score
• Performing cross_validation
  • using k-fold validation
    • used 10 samples
    • trained ,test this three classifiers

Knn and random forest performs the best.

• saving the model