Support the site!
You can also support LearningTensorFlow.com by becoming a patron at Patreon. If we have saved you trawling through heavy documentation, or given you a pointer on where to go next, help us to create new lessons and keep the site running.
I gave a talk at PyCon AU 2016 on TensorFlow, covering many of the first tutorials on this site.
This talk covers what TensorFlow is, why/when you should use it, and cover the basics surrounding Variables, Placeholders, and Custom Functions. Importantly, there are several use cases *not* focused on data analytics - TensorFlow is more than just a machine learning library!
On this page you will find a whole range of extra resources, split up by the topic you want to learn more about.
Official TensorFlow Documentation
The official documentation provided by Google is very Machine Learning focused (as most tutorials are at this stage). That said, they are quite good from that point of view, but they do assume quite a high level of machine learning knowledge.
You can find them at TensorFlow.org. On this site are tutorials, but also background information on how TensorFlow works “under the hood” and academic work.
Learning Data Mining with Python
If you don’t have a background in data mining, it can be quite hard to work through the existing tutorials. Robert Layton, the author of these tutorials, has also written a book introducing data mining concepts. The book is aimed at people that can already program, and examples are given in the Python programming language. While the book doesn’t cover TensorFlow, it does cover some related libraries. Click the book below to go to the Amazon page to purchase.
There are many great resources to learn the Python programming language more effectively. The official documentation is a great resources, and comes with several tutorials.
prettytensor Pretty Tensor provides a high level builder API for TensorFlow. It provides thin wrappers on Tensors so that you can easily build multi-layer neural networks.
tensorflow-white-paper-notes Annotated notes and summaries of the TensorFlow white paper, along with SVG figures and links to documentation
NTM-tensorflow Tensorflow implementation of Neural Turing Machine. This implementation uses an LSTM controller. NTM models with multiple read/write heads are supported.
tensorlayer is designed to provide a higher-level API to TensorFlow in order to speed-up experimentations and developments.
Theano and Lasagne are libraries that do similar things with computation as TensorFlow. Their documentation would provide more learning opportunity for TensorFlow – while the code is different, translating the code from Lasagne to TensorFlow would provide a great “extra resource”.
We have an increasing set of lessons that we hope guides you through learning this powerful library. Follow these links to keep going to our next lesson.
You can also use the nav menu at the top of the page to go directly to a specific lesson.
Coming soon (although not written by us):