TensorFlow 2.1.0 has been released, following two release candidates. The final version of the machine learning platform includes new features and breaking changes. Meanwhile, Python 2.7 has reached its end of life and is no longer supported by TensorFlow. Let’s take a look at what else has changed.
The machine learning platform TensorFlow, currently in version 2.0, is making its way toward the minor release 2.1.0: TensorFlow 2.1.0-rc0 is the first release candidate and includes some breaking changes. The upcoming version will be the last to support Python 2.7.
Google introduced TensorFlow Enterprise, a new collection of machine learning services and products. The beta version includes managed services, and some versions will receive long-term version support for up to three years. At the same time, Google unveiled the new website TensorBoard.dev.
At Machine Learning Conference 2019 in Munich, Christoph Henkelmann gave a talk about TensorFlow training on the JVM. We recorded the whole thing, and now you can watch it here (including slides) to learn all about how to combine a TensorFlow model with Java.
Every Monday, we take a step back and look at all the cool stuff that went down during the previous week. Last week we welcomed the new version of Keras, got one step closer to TensorFlow 2.0 with a new release candidate, and learned more about the potentials of blockchain.
We are one step closer to TensorFlow 2.0.0. The new release candidate, 2.0.0-rc2 includes new features, improvements, breaking changes, and bug fixes. Catch up on what’s expected to arrive in TensorFlow 2.0. The next major release focuses on ease of use and simplicity, with plenty of updates and easy model building with Keras.
Keras version 2.3.0 is here, and it is the last major multi-backend release. Going forward, users are recommended to switch their code over to tf.keras in TensorFlow 2.0. This release brings API changes and a few breaking changes. Have a look under the hood and see what it includes, as well as what the plans are going forward.
Machine learning has many use cases and offers up a world of possibilities. However, some people might be put off and think it’s too difficult. It’s not. You don’t have to be a machine learning pro to use TensorFlow Lite. Here’s how to get started building your own customized machine learning model on Android.
Every Monday, we take a step back and look at all the cool stuff that went down during the previous week. Last week, we had news coming from all direction with some announcements *cough* Quarkus *cough* stealing the spotlight!
Worried your ML models might blab about proprietary data? Now, developers can keep their training data isolated from their machine learning models with TensorFlow Privacy. This Python library optimizes ML models without running into any data security or privacy concerns with differential privacy.
Machine learning for mobile and Internet of Things devices just got easier. With the latest updates to TensorFlow Lite 1.0, ML heads towards your smart phone and smart home. See what new things the TensorFlow Dev Summit 2019 brings to the table.
Learning a language is difficult enough for humans; imagine how hard it must be for your neural network. Thanks to Lingvo, TensorFlow has a new framework for sequencing models for language tasks like machine translation, speech recognition, and speech synthesis. Now, who knows the word for “framework” in Esperanto?
TensorFlow 2.0 is on its way! What can we expect from this long-awaited upgrade to one of the most popular machine learning projects? A sneak peek at the preview version suggests a cleaner API, eager execution, and a tighter integration with tf.keras.
2019 is just around the corner so we’d like to find out which technologies will dominate next year, which technologies will fall behind and what’s going to stay static. We received over 350 votes so we humbly thank you for your participation.