At Man Group, we believe in the Python ecosystem and have been trading Machine Learning based systems since early 2014.
In June, we had the pleasure of welcoming Dani Papamaximou, Peter Bleackley, Tariq Rashid and Cheuk Ho to the PyData Meetup.
Dani Papamaximou on “Seismic analyses using python for civil engineering applications”
I am developing a python module to calculate the depth of the competent material below train tracks to accommodate high Rayleigh wave velocities. This module is using site investigation data and through a series of processes calculates the optimum thickness of the competent strata below the track of the train. This enables the contractor to have an early estimate of the imported material needed or inform the operating authority on the optimum speed of the train using a maintenance cost- profit analysis.
Peter Bleackley on “The Grammar of Truth and Lies - Using NLP to detect fake news”
I present a simple classifier that uses grammatical and stylistic features to distinguish between fake news and reliable sources with 90% accuracy.
Tariq Rashid on "d3fdgraph: interactive force-directed graphs in Jupyter"
Python has many options for data visualisation, including in the Jupyter notebook. However there didn’t seem to be a simple one for visualising graphs of related data. Tariq decided to write his own package which is now available via pip, and works on google’s colab notebook too. He’ll briefly share how easy it is to write plugins that work in the notebook and demonstrate how easy it is to use d3fdgraph.