It uses diamonds dataset from kaggle. The preprossing included converting categorical columns to Ordinal values. I have used Map here, but will be using Label Encoder and also dropping columns which wouldn't have elp in predicting values.
Then created a RandomForestReggressor with 250 estimators to get a 99.279% accuracy on the whole dataset.
I will try to create a small application where, a person can enter the features of the diamond and a possible price will be returned.