(Mostly Python) implementation of predictive analytics models and data science problems
Date Updated: 31 Aug, 2016
This repository contains Python code written by me to implement a wide range of problems/models in data science and predictive analytics. It currently contains the following directories:
-
"machine learning-coursera-andrew-ng" which is a collection of most of the programming assignments in Andrew Ng's course on Machine Learning from Coursera. It contains Python 2.7 code to solve these assignments.
-
"neural-networks-from-scratch" which is Python 2.7 code that I've written to implement a neural network from scratch. This model has then been used to predict handwritten digits over the MNIST database.
-
"dialoguesys/src" which contains code for a chatbot implemented using the seq2seq algorithm, written in Tensorflow. Most of the algorithm's code is inside the directory "models/seq2seq_model.py"
-
"predicting-age-health-mri-volumes" which contains code for classifying brain health status based on MRI scan. This uses scikit-learn, and most of the code is in the file "models.py".
The first two projects of these are my own solo undertaking, while the last two were done in teams (however all the code on here has been written by me and my contribution in both these teams has been fairly central).