Ultimate-housing-price-prediction

The main code content lies in the MAIN HOUSING jupyter notebook and the housing data stands in housing.csv

This is a ultimate project of housing price prediction , The entire data have been cleaned , filtered , scaled , trained and fine tuned rigorously over various levels . Data used for training : californina housing data No. of training examples : around 16 thousands. No. of test examples : around 4 thousands. Library/Dependencies : Numpy, Pandas , Matplotlib , sky-kit learn. encoder used : one hot encoder models trained : linear regressor , decision tree regressor , random forest regressor. Final model selected : random forest regressor Accuracy tested using cross validation (RMSE) and (r squared value) Grid search cv used for fine tuning of hyperparameters.

Upon the final training : RMSE scores : 48100.394269148856 , R square scores : 82.24797610379628 %

the overall performance of the model stands good and does not consider much over/underfitting issues (if any).

<-------Thanks for reading----------->