Convolutional Neural Networks Study Group - Third Meetup!
Details
Thank you to OpenSlate (https://openslatedata.com) for sponsoring and providing refreshments for our study group!!
The focus of this study group will be learning how to implement ConvNets aka Convolutional Neural Networks using TensorFlow and Keras.
We will not have an official textbook, but I personally recommend Hands On Machine Learning (https://www.safaribooksonline.com/library/view/hands-on-machine-learning/9781491962282/).
We will use the video lectures from Stanford CS231n: Convolutional Neural Networks (https://youtu.be/NfnWJUyUJYU?list=PLkt2uSq6rBVctENoVBg1TpCC7OQi31AlC) as our course material
Syllabus for CS231n from Spring 2017 (http://cs231n.stanford.edu/syllabus.html)
Topic of Study Group Discussion: Please watch lecture #4 from this list (https://www.youtube.com/playlist?list=PLkt2uSq6rBVctENoVBg1TpCC7OQi31AlC). Also look at the lecture #4 notes and slides from the link to the syllabus above.
We will ask that one volunteer each week take on the challenge of presenting their findings from the material to the rest of the group. The presentation should be approximately 10 minutes long.
Please also join our Slack group, you will find plenty of links of general interest there. We also have a channel set up called #stanford-cs231-cnns
https://join.slack.com/t/python-and-datasci/shared_invite/MjI0NDY4NzMwNTEyLTE1MDIzOTc0MTktODc0MTU4NmQ4YQ )
{The ultimate goal of this study group is to work through the first two problem sets from Stanford 231n over the course of the time of this study group(approx 7 sessions). (Problem Set #1 (http://cs231n.github.io/assignments2017/assignment1/), Problem Set #2 (http://cs231n.github.io/assignments2017/assignment2/))
Most of the problem sets do not need GPUs to run them, so feel free to set up a free account on https://notebooks.azure.com (https://notebooks.azure.com) or just run locally on a Jupyter notebook on your machine.}
Note: This study group will use Python. Working knowledge of Python would be helpful, but if you have a good understanding of at least one other programming language, Python should be fairly easy for you to pick up.