The project is about predicting the price of second hand cars. problem-Lack of knowlege about the current price of a used car in market often make it difficult for a person to buy or sell a car, solution- Many application and websites have made it easy for people to sell or buy cars, this model helps in finding the best market price of a car using machine learning algorithm with features like Year, Kilometers_Driven, new price and many more. Who can use it- people who want to check prices of a second hand car based on some required feature details. Dataset downloaded from kaggle, link- https://www.kaggle.com/avikasliwal/used-cars-price-prediction Features of the dataset are as follows- NUMERICAL FEATURES: year- In which year the car model was launched. Kilometers_Driven- No. of kilometers the car has been driven till date. Fuel_Type- type of fuel the car uses. Engine- cc of engine installed in the car. Power- power the car gives. Seats- the no. of seats. New_Price- showroom price of the same model currently. CATEGORICAL FEATURES: Transmission- car is automatic or manually driven. Owner_Type- Was sold how many times? Mileage- mileage of the car currently. Accuracy - 97.95%
Importing required libraries Importing dataset Data preprocessing Features Engineering: Filling the missing values which covers less than 5% of that feature Encoding categorical data Data cleaning Visualizing four numerical features with missing value based on their density for features with semantic graph the missing values are filled by mean of the presnt values whereas for features with sekewed graph the missing values are filled by median of the presnt values for categorical features getting dummies columns so that the features get converted to numrals in which the ML model can work ploting heatmap to visualize the corealtion between features. Building Model Scaling the dataset using standardScaler for coping up with the varied measuring units involved in the features. Applying various regression algorithms. As RandomForestRegressor shows good accuracy score , selecting the algorithm and applying hyperparameter tuning algorithm - GridSearchCV to it. Calculating the root-squared-error and mean-squared-error
The project is about predicting the price of second hand cars. 
