Swift for TensorFlow is a result of first-principles thinking applied to machine learning frameworks, and works quite differently than existing TensorFlow language bindings. Whereas prior solutions are designed within the constraints of what can be achieved by a (typically Python or Lua) library, Swift for TensorFlow is based on the belief that machine learning is important enough to deserve first-class language and compiler support.
Source: Swift Community | TensorFlow
What languages a developer needs to learn in order to traverse all the different domains that come together to deliver “full stack” applications is something I revisit every year in order to keep my skills and knowledge up-to-date. Always happy to have options that fold into my core skills 🙂
Nerd tip: Want to watch all those <developer conference of your choice> videos but don’t have 45 minutes or an hour at a time to spend on one of them, let alone dozens of them? Read on…
There are dozens of videos I always want to watch from Apple’s WWDC each year, but depending on the presenter, they can be painful to sit through. Lynda.com has a customized video player in their website that has ability to speed up the video while you are watching is without distorting the presenter’s voice too much, so I was hoping that I could find a way to watch Apple’s videos in the same manner.
If the videos you want to watch are available to download, (i.e. WWDC 2015 videos are as HD or SD .mp4’s) pull them down locally and play them in the QuickTime Player app. Once they start playing, hold down the option key (on a Mac anyways) and tap the fast forward button. It will speed up the video in increments of .1, all the way up to 2.0. The cool part is that it maintains the pitch of the audio, so you aren’t listening to chipmunks.
I find my ability to absorb what they are saying is pretty stable at 1.7. If they get into new material that requires some deep listening and processing, I’ll kick it back down to 1.4.
You’ll cover a lot more videos this way. Enjoy!