TensorFlow
Youtube search... ...Google search
- Google offerings
- Libraries & Frameworks
- TensorFlow is a Python library
- Keras (currently part of TensorFlow 2.0)
- Google launches TensorFlow 2.0 with tighter Keras integration | Khari Johnson
- TensorFlow.js ... Browser and Node Server
- TensorFlow Serving .. Cloud, On-prem ...support model versioning (for model updates with a rollback option) and multiple models (for experimentation via A/B testing)
- TensorFlow Lite ... Android, iOS, Raspberry Pi
- Converting to TensorFlow Lite convert models into TensorFlow Lite format
- TensorFlow Federated
- TensorFlow Extended (TFX) an end-to-end platform for deploying production ML pipelines
- TensorFlow Playground
- TensorBoard with Embedding Projector - visualization of high-dimensional data
- We’re making tools and resources available so that anyone can use technology to solve problems | Google AI
- TensorFlow Workshops click on links to run in Colaboratory (Colab)
- Machine Learning Crash Course with TensorFlow APIs | Google
- TensorFlow without a PhD | Martin Görner
- Running Tensorflow in Production | Matthias Feys
- Simple Tensorflow Cookbook
- TensorFlow-Course | GitHub
- TensorFlow Hub is a library for reusable machine learning modules that you can use to speed up the process of training your model. A TensorFlow module is a reusable piece of a TensorFlow graph. With transfer learning, you can use TensorFlow modules to preprocess input feature vectors, or you can incorporate a TensorFlow module into your model as a trainable layer. This can help you train your model faster, using a smaller dataset, while improving generalization.
- Neural Structured Learning (NSL) in TensorFlow
- Git - GitHub and GitLab
- The API...
- tf.estimator available alongside the newer Keras high-level API.
- tf.function a wrapper to use when writing certain functions in Python
- tf.Transform includes converting between formats, tokenizing and stemming text and forming vocabularies
- Related...
- Coding TensorFlow <--- video series
TensorFlow focuses on simplicity and ease of use, with updates like eager execution, intuitive higher-level APIs, and flexible model building on any platform, tight Keras integration. You can easily ingest datasets via tf.data pipelines, and you can monitor your training in TensorBoard directly from Colaboratory and Jupyter Notebooks. TensorFlow 2.0 and Google Cloud AI make it easy to train, deploy, and maintain scalable machine learning models | Paige Bailey and Barrett Williams - Google
- What’s coming in TensorFlow 2.0 | TensorFlow Team - Medium
- What’s in store for ML developers in TensorFlow 2.0? | Jane Elizabeth
Contents
GPU
tf.function
Eager Execution (Default in 2.0) =
tf.data - TF Input Pipeline
tf.distribute
mesh-TensorFlow
TensorFlow Probability
Reinforcement Learning with TF-Agents
Examples
Tensorflow for Practice Specialization
- Chapter 1:
- Presentation
- Meetup #1 fun-packed event
- Notes:
- Why map the pixel grayscale [0,1 to [0.01, 0.99] before feeding to the neural network?]
- How to Normalize, Center, and Standardize Image Pixels in Keras | Jason Brownlee
