I recently lead a 2.5hr ‘hardcore’ workshop at the DataScience SG MeetUp in Singapore.

The workshop was somewhat similar in structure to my previous PyCon one - except that it had a DataScience flavour. To add to that, I created a new ‘application’ that showed how Autoencoders could be used to do anomaly detection (with MNIST as the base data).

In addition to that, we went through the CNN-Commerce example, so that people could see how a pretrained CNN could be repurposed to be used on classes that hadn’t previously been encountered by the network.

PS: And if you liked the Workshop, please ‘star’ the Deep Learning Workshop repo :: Star

Presentation Screenshot

If there are any questions about the workshop please ask below, or contact me using the details given on the slides themselves.

Presentation Content Example

And the excellent folks at Engineers.sg also generously made a video of the presentation which is available on their site. Their videos are normally excellent, though I can’t vouch for this one, since I can’t bear to watch myself…



Martin Andrews

{Finance, Software, AI} entrepreneur, living in Singapore with my family.



blog comments powered by Disqus