Published on

TensorFlow 2.0 : CNNs for Images


Sam Witteveen and I started the TensorFlow and Deep Learning Singapore group on MeetUp in February 2017, and the twenty-fourth MeetUp, aka TensorFlow and Deep Learning : Images and CNNs, was again hosted by Google Singapore.

My talk was titled "First steps in Deep Learning with TensorFlow 2.0 : CNNs", and was intended to introduce the elements of a CNN model before explaining the intricacies of back-propagtion (i.e. in some ways it was a different approach from the typical "build from the ground up" NN introductory pitch).

Timothy Liu talked about how to use pretrained models (including from tf.keras.applications), and Sam explored how Progressively growing training works for ResNet-50 (from scratch).

The slides for my talk (as a 400Kb PDF) are here :

Presentation Screenshot

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

Presentation Content Example