Data is present everywhere at malls, multiplexes, online shopping sites, dating sites, government intelligence almost everywhere that continuously needs to be monitor & analyse, which is very huge data, which cant be analysed by human, at that point machine learning came into picture.
For introduction to that #PythonPune added new chapter to its Pythonpune Meetup which was held on 30 April 2016 at Red Hat office, Magarpatta.
Almost 70-80 peoples were there(which was unexpected for us). :P
Very few peoples which were present who know machine learning. Rest of people doesn’t know but interested to learn . I was also interested to learn machine learning after watching blockbuster “person of Interest”
Speakers of meetup were Satish Patil and Sudarshan Gadhave who gave Introduction on Machine Learning. Satish is a founder and chief data scientist of Lemoxo Technologies, where he advises companies large and small on their data strategy. He holds a Ph.D. from the University of Minnesota, USA. Over the last 4 years, Satish has helped global pharma, e-commerce giants & startups to solve complex data problems to drive the meaningful business outcome. His core competency lies in applied math and statistical modelling, machine learning, deep learning and data visualisation. Satish is passionate about applying math, technology, design thinking and cognitive science to better understand, predict and improve business functions. Sudarshan works @ NEC as data scientist.
Machine learning is all about technical,business and statistics. In this session, we learnt basics of machine learning and how to implement machine learning algorithms on your data sets using Python and Scikit-Learn. also we learnt: The Black Box of Machine Learning,features,tupules,training and test data set,classification,clustering,pure & impure states,entropy,decision tree, supervised and unsupervised learning, market basket analysis, data pre-processing, K means algorithm. we have used titanic data sets for example (https://github.com/pcsanwald/kaggle-titanic/blob/master/train.csv).
Required tools/libraries for machine learning :
ChandanKumar talked about Fedora Labs. Fedora labs have the selection of curated bundles of purpose-driven software and content as curated and maintained by members of the Fedora Community. These may be installed as standalone full versions of Fedora or as add-ons to existing Fedora installations. From fedora-labs, we were focused on Fedora Scientific (open source software computing) which comes with featured application like ipython, pandas, Gnuplot, Matplotlib, R , latex, etc can be useful for machine learning.
To download fedora Scientific, Click Here.
If you need any help, you can ping on #fedora-science channel on Freenode in IRC.
Still from Meetup :
Satish Patil delivering Session on Machine Learning:
Sudarshan Gadhave Talking about Machine Learning:
ChandanKumar talking about Fedora Scientific and OpenSource Contribution: