- Here i have 8218 records of cars and it's features in which 2040 are unique cars
- Here i have trained various model using hyperparameter tunning and it took 80.42 minutes on intel i3 processor
- After training a model i got Gradient Boost as best model for this problem statement
- Gradient Boost algorithm gives 0.977921 r2_score it means 97.79% accuracy on training data set and 0.986117 r2_score it means 98.61% accuracy on testing data set, here data set was split on 80:20 ratio
- Here i have developed end to end application using Flask, Javascript, Bootstrap, CSS and HTML
- I have used google-images-download library to scrap links of images from google images.
- In this project i have used cars dataset from kaggle, you can get it from here
- I have deployed this on AWS Elastic Beanstalk platform Link: http://carvaluationprediction-env-1.eba-pvpbk242.us-east-2.elasticbeanstalk.com/
- Select any car fill the inputs with proper information then click on predict button you will get predicted price of that car.
- Create .ebextensions folder to your main directory
- Inside .ebextensions folder create a python.config file and write configration like this
- Create .ebignore file inside main directory
- Open the Elastic Beanstalk console using this link: https://console.aws.amazon.com/elasticbeanstalk/home#/gettingStarted?applicationName=getting-started-app
- Enter your application name.
- Application tags are optional so just ignore it.
- For Platform, choose a python platform.
- For Application code choose Upload your code.
- Upload a zip file of your project.
- Click on create application button.
- rest of things will take care by aws elastic bean stalk and you will get deployed link.
 |
|---|
| Jay Soni |
- Entire credits goes to My God
- Car images credits goes to google-images-download and google images search engine
- If you like my work and it helped you in anyway then please do ⭐ the repository it will motivate me to make more amazing projects