geooot / tamudatathon2019

ConocoPhillips TAMU datathon challenge

ConocoPhillips Datathon Challenge

[Jupyter Notebook] [Kaggle Competition] [TAMU Datathon]

ConocoPhillips give a challenge for The TAMU Datathon to make a predictor for whether an oil rig would fail given a set of sensor values (107 columns).

Here is our naive exploration of creating various classifiers to predict of a given state of sensor data would lead to a rig failure.

Key takeaways

• Pandas is pretty neat
• Jupyter is also pretty neat
• We live in the matrix where Ax = b
• We have a long way to go to be data science masters

Authors

• Aditya Pethe
• Anikait Sharma
• George Thayamkery
• Jon Waterman