machine_learning_demo

Demonstration of my data science skills (linear regression, random forest, neural network, bayesian, deep learning) using python ecosystem for machine learning (pandas, numpy, tensorflow, scikit-learn)

Pierre-Edouard GUERIN

---

Prerequisites

• To run these analysis with python on a local machine using linux terminal, see INSTALL.sh.
• To run on binder server using already built docker image with all prerequisites, simply click on the button 'launch binder' related to the analysis

Contents

Machine Learning

(Scikit-learn, Pandas, Seaborn, Matplotlib, Xgboost, Lightgbm)

• Linear regression: Predicting Housing Prices: A model to predict the value of a given house in the Boston real estate market using linear regression improved by tree-like modeling methods. Identified the best price that a client can sell their house utilizing machine learning.
  • run this analysis:
  • see the data: https://www.kaggle.com/c/house-prices-advanced-regression-techniques/data

Deep learning

(numpy, Keras, Tensorflow)

• Image Classification: Designing and implementing a Convolutional Neural Network that learns to recognize Simpson characters
  • run this analysis:
  • see the data: https://www.kaggle.com/alexattia/the-simpsons-characters-dataset