AutoML Car Valuation

• 

  Unlock the value of used cars: a machine learning project

Inspiration

As car enthusiasts, we were interested in building a machine learning model that could predict the price of a used car based on its make, model, year of manufacture, mileage, and other features. We found a great dataset on Kaggle that contained information about thousands of used cars, and decided to use it to train our model.

What it does

Our model takes in various features of a used car and predicts its selling price. We used a deep learning approach and built a neural network using TensorFlow and Keras.

How we built it

We began by cleaning and preprocessing the data, which included dropping unnecessary columns, converting categorical variables to binary, and one-hot encoding. We then split the data into training and testing sets, and used feature scaling to normalize the data. We built a neural network using Keras, with three dense layers, and trained it on the training set. We then evaluated the model on the test set and saved it for future use.

Challenges we ran into

One of the challenges we faced was deciding which features to include in our model. We had to consider which features were most important in determining the selling price of a used car. Additionally, we had to fine-tune the hyperparameters of our model to ensure that it was accurate and efficient.

Accomplishments that we're proud of

We're proud of building a deep learning model that can accurately predict the selling price of a used car based on its features. We were able to achieve an R-squared score of over 0.9, which indicates that our model can explain a high percentage of the variance in the data.

What we learned

Through this project, we learned how to clean and preprocess data for machine learning, how to build and train a neural network using Keras, and how to evaluate the performance of a machine learning model. We also gained a better understanding of the factors that affect the selling price of a used car.

What's next for PREDICTING THE PRICE OF THE OLD CAR

In the future, we plan to incorporate additional features into our model, such as the car's location and the number of previous owners. We also hope to optimize our model further to improve its accuracy and efficiency. Ultimately, we hope that our model can be used to help people make informed decisions when buying or selling a used car.

Built With

• django
• github
• jupyter
• kaggle
• keras
• notebook
• numpy
• pandas
• python
• scikit-learn
• tensorflow