TensorFlow 2.7 is released
(10 November 2021)

Other recent news...

Microsoft highlights .NET 8 Hardware Intrinsics features
Microsoft highlights .NET 8 Hardware Intrinsics features
Additional hardware functionality via APIs so the bespoke web apps we develop can take advantage of users' underlying hardware, e.g. WebAssembly support so that core algorithms can...
11 December 2023
Google launches Gemini AI model
Google launches Gemini AI model
Gemini was designed for flexibility, so it can run on everything from data centres to mobile devices. It's been optimized for three different sizes: Ultra, Pro and Nano (for on-dev...
6 December 2023
Apple launches release candidate versions of iOS 17.2 & iPadOS 17.2
Apple launches release candidate versions of iOS 17.2 & iPadOS 17.2
Developing and testing our bespoke Apple iOS apps on beta versions of iOS 17.2 & iPadOS 17.2 allows our app developers to confirm that the apps we develop will work as expected...
5 December 2023
TensorFlow 2.7 is released
This TensorFlow release improves usability with clearer error messages, simplified stack traces, and adds new tools and documentation for users migrating from TensorFlow 1 to TensorFlow 2.

Improved Debugging Experience
The process of debugging your code is a fundamental part of the user experience of a machine learning framework. In this release, we've considerably improved the TensorFlow debugging experience to make it more productive and more enjoyable, via three major changes: simplified stack traces, displaying additional context information in errors that originate from custom Keras layers, and a wide-ranging audit of all error messages in Keras and TensorFlow.

Simplified stack traces
TensorFlow is now filtering by default the stack traces displayed upon error to hide any frame that originates from TensorFlow-internal code, and keep the information focused on what matters to you: your own code. This makes stack traces simpler and shorter, and it makes it easier to understand and fix the problems in your code.

Automatic context injection for Keras layer exceptions
One of the most common use cases for writing low-level code is creating custom Keras layers, so we wanted to make debugging your layers as easy and productive as possible. The first thing you do when you're debugging a layer is to print the shapes and dtypes of its inputs, as well the value of its training and mask arguments. We now add this information automatically to all stack traces that originate from custom Keras layers.

Audit and improve all error messages in the TensorFlow and Keras codebases
Lastly, we've audited every error message in the Keras and TensorFlow codebases (thousands of error locations!) and improved them to make sure they follow UX best practices. A good error message should tell you what the framework expected, what you did that didn't match the framework's expectations, and should provide tips to fix the problem.

TensorFlow is an end-to-end open source machine learning platform used across a range of tasks but with a particular focus on training deep neural networks.

TensorFlow was first developed by the Google Brain team for internal Google use and was later released under an Apache license in 2015. A major updated version of TensorFlow, named TensorFlow 2.0, was released in September 2019.

TensorFlow serves as a core platform and library for machine learning. TensorFlow’s APIs use Keras to allow users to make their own machine learning models.

Whilst TensorFlow provides a stable API in the Python programming language, third-party language binding packages are also available for other programming languages such as C# which has been used by New Media Aid's app development team for developing .Net web apps since it was first released back in 2001. You can see a simple example of image recognition using TensorFlow Lite by uploading an image on this page.

TensorFlow is the core open source library helps New Media Aid's app engineers to develop and train ML models.

TensorFlow.js is a JavaScript library for training and deploying models in the browser and on Node.js.

TensorFlow Lite is a lightweight library for deploying models on mobile and embedded devices.

TensorFlow Extended is an end-to-end platform for preparing data, training, validating, and deploying models in large production environments.

Why TensorFlow?
Whether you're an expert or a beginner, TensorFlow is an end-to-end platform that makes it easy for you to build and deploy ML models.

Easy model building
TensorFlow offers multiple levels of abstraction so you can choose the right one for your needs. Build and train models by using the high-level Keras API, which makes getting started with TensorFlow and machine learning easy. If you need more flexibility, eager execution allows for immediate iteration and intuitive debugging. For large ML training tasks, use the Distribution Strategy API for distributed training on different hardware configurations without changing the model definition.

Robust ML production anywhere
TensorFlow has always provided a direct path to production. Whether it's on servers, edge devices, or the web, TensorFlow lets you train and deploy your model easily, no matter what language or platform you use. Use TensorFlow Extended (TFX) if you need a full production ML pipeline. For running inference on mobile and edge devices, use TensorFlow Lite. Train and deploy models in JavaScript environments using TensorFlow.js.

Powerful experimentation for research
Build and train state-of-the-art models without sacrificing speed or performance. TensorFlow gives you the flexibility and control with features like the Keras Functional API and Model Subclassing API for creation of complex topologies. For easy prototyping and fast debugging, use eager execution. TensorFlow also supports an ecosystem of powerful add-on libraries and models to experiment with, including Ragged Tensors, TensorFlow Probability, Tensor2Tensor and BERT.

Intro to ML
Machine learning is the practice of helping software perform a task without explicit programming or rules. With traditional computer programming, a programmer specifies rules that the computer should use. ML requires a different mindset, though. Real-world ML focuses far more on data analysis than coding. Programmers provide a set of examples and the computer learns patterns from the data. You can think of machine learning as “programming with data”.

Steps to solving an ML problem
There are multiple steps in the process of getting answers from data using ML. For a step-by-step overview, check out this guide that shows the complete workflow for text classification, and describes important steps like collecting a dataset, and training and evaluating a model with TensorFlow.

Anatomy of a neural network
A neural network is a type of model that can be trained to recognize patterns. It is composed of layers, including input and output layers, and at least one hidden layer. Neurons in each layer learn increasingly abstract representations of the data. For example, in this visual diagram we see neurons detecting lines, shapes, and textures. These representations (or learned features) make it possible to classify the data.

Training a neural network
Neural networks are trained by gradient descent. The weights in each layer begin with random values, and these are iteratively improved over time to make the network more accurate. A loss function is used to quantify how inaccurate the network is, and a procedure called backpropagation is used to determine whether each weight should be increased, or decreased, to reduce the loss.

TensorFlow Hub is a repository for machine learning models.
From image classification, text embeddings, audio, and video action recognition, TensorFlow Hub is a space where our app developers can browse trained models and datasets from across the TensorFlow ecosystem. The hub allows our app engineers to:
  • Find trained models for transfer learning to save time on training
  • Publish your own models
  • Deploy models on device and in the browser
The four areas of machine learning education
It is important for New Media Aid's app engineers / app developers to understand how to learn machine learning (ML). The learning process breaks down into the four areas shown below:
  • Coding skills: Building ML models involves much more than just knowing ML concepts—it requires coding in order to do the data management, parameter tuning, and parsing results needed to test and optimize your model.
  • Maths and stats: ML is a maths heavy discipline, so if our app developers plan to modify ML models or build new ones from scratch, familiarity with the underlying math concepts is crucial to the process.
  • ML theory: Knowing the basics of ML theory will give you a foundation to build on, and help you troubleshoot when something goes wrong.
  • Build your own projects: Getting hands on experience with ML is the best way to put your knowledge to the test.
more news...