addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramlinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Introducing Python for Data Science (scikit-learn, Pandas, Kaggle, Cython and mo

Ian Ozsvald & Bart Baddeley will demo four areas of data science using Python. The goal is to introduce a number of these areas that may not be familiar to more traditional Python developers, this could open into an extended future session if there's interest.


The talk will cover:

• Using the IPython Notebook and Scikit-learn to begin solving the Kaggle Titanic machine learning problem

• Natural Language Processing and Machine Learning for social media analysis using scikit-learn

• Using Pandas and QGIS to solve the Future Cities London hackweekend challenge

• High Performance Python using profiling techniques, Cython and new tools like Numba to calculate at C-like speeds


** Skills Matter is kindly providing host to this talk so please register at: http://skillsmatter.com/event-details/home/introducing-python-for-data-science-scikit-learn-pandas-kaggle-cython-and-more


Join or login to comment.

  • Ian O.

    Here's a short write-up of the talk with links to some of the presentations, along with the video link: http://ianozsvald.com/2013/11/01/introducing-python-for-data-science-talk-at-skillsmatter/

    3 · November 1, 2013

    • Luciano J.

      Thanks for the recording!

      November 1, 2013

  • Seref A.

    Thanks to both speakers! Especially for the clues about Cython.

    November 1, 2013

  • aaron

    Is there an online sign-up toregister interest for the workshops that were mentioned by Ian?

    October 31, 2013

    • Alberto C.

      Which workshops are those? (for the people who could not assist, like me)...

      October 31, 2013

    • Ian O.

      Hi Alberto. I'm thinking of running some training courses in the new year on both Data Science with Python and High Performance Python - send me an email (ian at ianozsvald com) if you're interested and I'll add you to the list.

      November 1, 2013

  • Alberto C.

    Is it the recording available finally?

    October 31, 2013

  • Dane W.

    Brilliant talks - the best presented and most interesting of all the Python talks I've been to this year. Would happily go to them all over again!

    October 31, 2013

  • Neil G.

    Sad to have missed out. My company are hiring for new developers so if anyone fancies a contract or perm role let me know. We are using OpenERP. Wandsworth based.

    October 31, 2013

  • Bryan F.

    Interesting talk, I particularly liked the worked out examples in Pandas. Contrary to what was said however scikit-learn could scale to billions of data points. You just use the stochastic gradient descent (SGD) family of classifiers. Essentially this initialises your model parameters to random values, and then you feed the classifier your dataset in small batches which amend that initial guess. Batches can be as small as one record at a time, though larger batches work better.

    Of course you'll probably get perfectly good results with a smaller random sample of the dataset: you may not actually need billions of data points to train your model. I'd have mentioned it at the time except I know more about machine learning than scikit-learn, and had to check out scikit's website to see if it supported this :)

    October 30, 2013

    • Ian O.

      Cool, I didn't know about that. Thanks for the pointer.

      October 31, 2013

  • Andrey D.

    Very sorry, but can't attend. :(

    October 30, 2013

  • Orlando H.

    Planning to come along...

    October 30, 2013

  • Colin M.

    Have to cry off due to work....

    October 30, 2013

  • Ian H.

    Unfortunately work commitments mean I can't make it.

    October 30, 2013

  • Ian O.

    I'm humbled by the large sign-up (I'm one of the speakers)! It'll be good to meet you all tomorrow. There are related posts on my blog if any of you want some background reading.

    1 · October 29, 2013

    • Ian O.

      October 29, 2013

    • Emlyn C.

      You can't be that surprised Ian! :) Python is the path of least resistance for most people once they've hit the "Excel Wall". It's an easier syntax than VBA, not to mention more capable and performant.

      October 30, 2013

  • Sachin T.

    I will be attending

    October 30, 2013

  • Venera M.

    We have discovered a lot of requests for Python developers on our platform — simple2connect iPhone app. Couple of London startups have opportunities in this field right now.

    October 30, 2013

  • Luciano J.

    Gutted I won't make it! I have another seminar at the same time.

    Will this be made available as a recording online? I saw that the last python group presentation ("TDD for everyone") was being recorded while I was there .... Ian and Bart, are you planning on doing a re-run or follow up for this sometime?

    October 29, 2013

    • Antonio C.

      There will be recording for the talk but I think we can ask to edit that bit

      October 30, 2013

    • Ian O.

      I think it'll be ok to record it all with no editing, I can edit my slides after if we get far into the solution (but in 15 mins...I doubt we will!).

      October 30, 2013

  • Alberto C.

    Unfortunately I won't be able to assist! :-(

    Last-minute meeting request for tomorrow at 4pm makes it impossible for me, as I'd come from Southampton.

    Would be great if the workshop is recorded!!

    October 29, 2013

  • aaron

    Hopefully i f I can finish work on time

    October 24, 2013

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy