Going Deeper :
Transfer Learning

TensorFlow & Deep Learning SG

Martin Andrews @ redcatlabs.com

16 February 2017

About Me

  • Machine Intelligence / Startups / Finance
    • After PhD, went into Finance as a Quant
    • Moved from NYC to Singapore in Sep-2013
  • 2014 = 'fun' :
    • Machine Learning, Deep Learning, NLP
  • Since 2015 = 'serious' :: NLP + deep learning
    • SG funded project
    • & Papers...

Overview

"Something more challenging"

  • Use a state-of-the-art TensorFlow model
  • Solve a problem it wasn't trained on
  • ... Deep Learning as a component

Goals

  • Distinguish between Classic and Modern sports cars
    • Could be other types of images
    • Could be many different classes
  • Very small training time
  • Very small number of training examples
  • Able to be put into production

Image Classification

  • ImageNet Competition
  • over 15 million labeled high-resolution images...
  • ... in over 22,000 categories

ImageNet Karpathy

2014 : State-of-the-Art

Google Inception-v1

GoogLeNet - aka Inception v1

Approach

  • Take an existing model trained for ImageNet
  • Instead of softmax layer :
    • use the logits as features to train an SVM
  • Support Vector Machines are 'well understood' tools

Network Picture

Repurposed Network

The Notebook

  • Code is in GitHub

Conclusions

  • Really works
  • Didn't have to train a deep neural network
  • Could plug this TF model into existing pipeline
GitHub - mdda

* Please add a star... *

- QUESTIONS -


Martin.Andrews @
RedCatLabs.com


My blog : http://mdda.net/

GitHub : mdda

Our Next MeetUp

  • "TensorFlow and Deep Learning Singapore"
    • Hosted by Google
  • 20-March-2017 : Monday evening
    • Something for beginners
    • Something more challenging
    • Lightning talks

TensorFlow CodeLab

  • 4-March-2017 : Saturday 9:30am - 1:00pm
  • Held here at Google
  • Targeting TensorFlow beginners
  • Material : TensorFlow for Poets

10-week Deep Learning
Developer Course

  • Plan : Mid-May to End-July
  • Weekly 3-hour sessions will include :
    • Instruction
    • 4 structured projects
    • 2 self-directed projects
  • Cost: S$TBD
  • Expect to work hard...