Skip to content

Latest commit

 

History

History
10 lines (6 loc) · 608 Bytes

File metadata and controls

10 lines (6 loc) · 608 Bytes

Kaggle-House-Pirces-With-Tensorfow

An optimized Neural Network for the Kaggle Competition: https://www.kaggle.com/c/house-prices-advanced-regression-techniques/data

A neural network program using Tensorflow that estimates the price of a house. Works off data from the afformentioned kaggle competition.

The structure is a multilayerd perceptron that uses one-hot encoding. Uses the leaky_relu activation function; however, this can be changed in the layer_activation variable.

Takes data from file train.csv with 81 feilds.

Can export a tranined network or import a network under the name "network".