Traffic Sign Recognition with TensorFlow Deep Learning

Date

Monday, May 22, 2017 - 6:30pm - 8:30pm

Venue

Pivotal

Pivotal
3495 Deer Creek Road
Palo Alto, CA 95124

Bring ID (e.g. Driver's Licences) for Pivotal Security

Speaker: 
Waleed Abdulla

Event Details

Event Details  
1) Go through the simplest possible model (1 neuron per traffic sign) and then iteratively improve it, showing the effect of each improvement on the accuracy of the neural network. 
For background, read part 1 of his related blog post before the meetup: 
https://medium.com/@waleedka/traffic-sign-recognition-with-tensorflow-629dffc391a6 This has a Jupyter notebook using Python 3.5 and TensorFlow 0.12. There is also a Docker image if you prefer.

2) Cover multi-layer models and convolutional networks, to get near human level accuracy.

Speaker Bio

Entrepreneur, software engineer, deep learning consultant. 
https://github.com/waleedka 
https://www.linkedin.com/in/waleedka/

Event page provided by ACM