Repository files navigation

A neural network with flexible parameters - number of layers, number of neurons, regularization lambda value etc.

Flexnet uses data from several sources to induce different models.

https://archive.ics.uci.edu/ml/datasets/Haberman%27s+Survival

https://archive.ics.uci.edu/ml/datasets/wine

https://archive.ics.uci.edu/ml/datasets/Contraceptive+Method+Choice

https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic)

Flexnet measures accuracy and other quality indicators with cross validation, selecting the best models among the a pool of differently parameterized networks.

This is a project for UFRGS's (Univerdade Federal do Rio Grande do Sul) subject Neural Networks and Fuzzy Systems, of the Computer Engineering degree.