Neural network development and training using TensorFlow DNNRegressor class

This program develops a neural network to predict median_house_value in the california housing dataset (https://download.mlcc.google.com/mledu-datasets/california_housing_train.csv) using synthetic features. Different parameteres of the model are adjusted to minimize the root mean squared error (RMSE) and the model is then tested on california housing test set (https://download.mlcc.google.com/mledu-datasets/california_housing_test.csv) and gives an RMSE of 99.067.

Update:

Trained and tested the same ANN on heart dataset to classify whether the patient is positive or negative for the heart disease. The model gave an RMSE of only 0.1318.

Tools and Technologies used: Python, TensorFlow, Pandas, Numpy