This repository contains some Python programs that implement different Deep Learning techniques. Some of the program has been done following the online course on udemy Deep Learning A-Z™: Hands-On Artificial Neural Networks:
https://www.udemy.com/deeplearning/
In particular, right now the following programs are present:
-
ANN Clients Bank:
The programs is implemented in Python 3 using Tensorflow and Keras. The main goal is customer churn prediction in a Bank. The Hyperparameters optimization is done using GridSearch
-
Cat-Dog CNN:
The programs is implemented in Python 3 using Tensorflow and Keras. We implemented a CNN able to recognize between Cats or Dogs in a picture
-
LSTM Stock Price: The programs is implemented in Python 3 using Tensorflow and Keras. We implemented a Long Short Term Memory(LSTM) for Google Stock Price Prediction
-
SOM:
The programs is implemented in Python 3 using MiniSom implementation. The program use Australian Credit Approval data to find frauds thanks using Self Organizing Maps
-
SOM-ANN Fraud Detection:
The programs is implemented in Python 3 using MiniSom implementation, Tensorlfow and Keras. The program is an extension of SOM.py. It uses Australian Credit Approval data to find frauds thanks using Self Organizing Maps and then calculate the fraud probability using ANN.
