In this project we will predict the sale prices of homes in KingCounty, USA, between May 2014 and May 2015 By using deep learning techniques and several relevant features such as the number of bedrooms, bathrooms, view, and square footage, this project aims to develop a predictive model that can accurately estimate home prices in King County. Such a model can be used by real estate agents, buyers, and sellers to make more informed decisions about home pricing, purchase, and sale.
TRAIN A DEEP LEARNING MODEL WITH LIMITED NUMBER OF FEATURES once the model is trained(using train_set) we freeze those weights between the neurons and access the performance of network on completey new data set that model has not seen(test_set) this is the idea we make sure that thiss deep neural network are able to generalize not memorize the traning data
EVALUATE TRAINED DEEP LEARNING MODEL PERFORMANCE TRAIN AND EVALUATE A DEEP LEARNING MODEL WITH INCREASED NUMBER OF FEATURES (INDEPENDANT VARIABLES)