Predicting houses prices with ML models

Description about the training data:

License CC BY-NC-SA 4.0

Melbourne real estate dataset, created and cleaned by Tony Pinowell. It includes Address, Type of Real estate, Suburb, Method of Selling, Rooms, Price, Real Estate Agent, Date of Sale and distance from C.B.D.

First attempt

with In-sample decision tree model

this is found on the In-sample.py file

Second attempt

With practical model by spliting training and validation data

This is found on the model.py file

Error = 240496

Third attempt

Determine best tree size and train the model again

Treesize = 500 Error = 236940

Fourth attempt

Using a Pipline to impute missing numerical values and cross-validation /it{improvemore.py}

Using different numbers of estimators and ploting the scores