TensorFlow, Kubernetes, and more in this exciting episode!
It’s been a while guys! I’ve been pretty busy because there’s a lot going on, from giving a talk at TensorFlow Dev Summit Extended Manila 2018 to starting new projects.
So I’m going to release two episodes side-by-side! I hope you’re ready! 😉
The Save PDF to iBooks share option in Safari in iOS devices are super awesome! It was unexpectedly good! I use it to save webpages like guides, tutorials, and cheatsheets as PDFs, especially since I’m not online a lot. If you have an iOS device, it’s worth a try. If you don’t have an iOS device, you can use SimplePrint.
This year’s TensorFlow Dev Summit is awesome as usual! A lot of new tools and updates have been released and I’m so excited to try them out. We’ll look at two of my favorites next but if you want to learn more, here’s a link to the TensorFlow blog.
TensorFlow.js is one of the awesome new members of the TensorFlow family. TensorFlow.js has support for the TensorFlow Layers API, is hardware-accelerated with access to the low-level Ops API (formerly Deeplearn.js), can train models in-browser, and use exported Keras models & TensorFlow SavedModels.
Similar to how TensorFlow.js can import Keras and TensorFlow models, we can now import powerful machine learning models without breaking a sweat in a single line of code with TensorFlow Hub. TensorFlow hub is like a repository of popular ML models like Inception and the like. Heck, even NASNet is there.
A.K.A “Command Line Interactive Controller for Kubernetes”, Click helps Kubernetes users do their tasks faster by giving context to commands.
Crunch is a macOS application that compresses PNGs really well, but is “insanely” slow. You can give it a try if you want extra good compression and you have the time to do it.
PopMotion POSE is the declarative version of PopMotion. Declarative basically means you define what something should do, instead of being concerned with how and when it should happen. This allows you to have the power of PopMotion and giving it simplicity. This makes it really awesome to use with React, which is also declarative in nature.
Jenkins X brings the power and flexibility of Jenkins for CI/CD and bringing it to Kubernetes. While having a fair level of abstraction, it still hasn’t lost its flexibility, as you still can go deep and tailor everything to your liking.