Here's my own list of the interesting stuff announced during this year's WWDC, collected from the keynotes, various Apple docs, blog posts and tweets.
If you're planning to watch the videos, I really recommend this Mac app that helps you download and watch them: https://github.com/insidegui/WWDC.
http://www.apple.com/osx/elcapitan-preview/
Following the tradition from last year, here's my complete list of all interesting features and updates I could find in Apple's OSes, SDKs and developer tools that were announced at this year's WWDC. This is based on the keynotes, the "What's New In ..." presentations and some others, Apple's release notes, and blog posts and tweets that I came across in the last few weeks.
If for some reason you haven't watched the talks yet, I really recommend watching at least the "State of the Union" and the "What's New In" intros for the platforms you're interested in. The unofficial WWDC Mac app is great way to download the videos and keep track of what you've already watched.
If you're interested, here are my WWDC 2015 notes (might be useful if you're planning to drop support for iOS 8 now and start using some iOS 9 APIs).
This is a list of hacks gathered primarily from prior experiences as well as online sources (most notably Stanford's CS231n course notes) on how to troubleshoot the performance of a convolutional neural network . We will focus mainly on supervised learning using deep neural networks. While this guide assumes the user is coding in Python3.6 using tensorflow (TF), it can still be helpful as a language agnostic guide.
Suppose we are given a convolutional neural network to train and evaluate and assume the evaluation results are worse than expected. The following are steps to troubleshoot and potentially improve performance. The first section corresponds to must-do's and generally good practices before you start troubleshooting. Every subsequent section header corresponds to a problem and the section is devoted to solving it. The sections are ordered to reflect "more common" issues first and under each header the "most-eas
