How to predict housing price using Neural Network Toolbox?
2 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
I would like to predict housing prices using inputs such as distance away from subway, numbers of rooms etc.. which of the 4 wizards ( fitting tool, pattern recognition tool, clustering tool, time series tool) should i use?
Thank You WT Lim
0 comentarios
Respuesta aceptada
Greg Heath
el 10 de Oct. de 2014
Use the fitting tool with fitnet
Also, see
help house_dataset
doc house_dataset
help fitnet
doc fitnet
[ x,t ] = house_dataset;
[ I N ] = size(x) % [ 13 506 ]
[ O N ] = size(t) % [ 1 506 ]
MSE00 = mean(var(t',1)) % 84.42
net = fitnet;
rng('default')
[net tr y e ] = train(net,x,t); % e=t-y
NMSE = mse(e)/MSE00 % normalized MSE = 0.071101
R2 = 1-NMSE % Rsquare (See Wikipedia) = 0.9289
% ~ 93% of the target variance is modeled by the net. % Obtain details from the training record
tr = tr
Hope this helps
Thank you for formally accepting my answer
Greg
2 comentarios
Greg Heath
el 10 de Oct. de 2014
Better solutions probably can be obtained by using
1. A different number of hidden nodes
2. A different set of random initial weigts
Más respuestas (0)
Ver también
Productos
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!