A python program that uses Neural Network to predict strength of given samples of concrete.
This program uses a Neural Network of 3 layers (1 hidden layer) to predict strength of concrete. The following python libraries have been used:
- Pandas - for reading data into a dataframe.
- Numpy - for algebraic functions and handling data.
- sklearn - for using preprocessing package(for feature normalization), model selection package(for splitting data into training and testing parts) and exception package(for ignoring Data Conversion Warnings .
- scipy - for using optimize package for minimising cost function
The number of neurons in hidden layer and regularisation constant can be changed to see their effects on Accuracy of the NN.