TensorFlow is currently the hottest open-source project, widely used for numerical computation based on data flow graphs. As TensorFlow is widely applied, many derivative applications have emerged, and some believe that TensorFlow is the best way to build deep learning models today. This article compiles a list of excellent practices, libraries, and projects related to TensorFlow, covering six themes of outstanding resources.
1. Tutorials
TensorFlow Tutorial 1
https://github.com/pkmital/tensorflow_tutorials
From basics to more interesting TensorFlow applications
TensorFlow Tutorial 2
https://github.com/nlintz/TensorFlow-Tutorials
An introduction to deep learning based on the Google TensorFlow framework, these tutorials are direct ports of Newmu’s Theano
TensorFlow Examples
https://github.com/aymericdamien/TensorFlow-Examples
Tutorials and code examples for beginners in TensorFlow
Sungjoon’s TensorFlow-101
https://github.com/sjchoi86/Tensorflow-101
TensorFlow tutorials written using Jupyter Notebook in Python
Terry Um’s TensorFlow Exercises
https://github.com/terryum/TensorFlow_Exercises
Recreating code from other TensorFlow examples
Installing TensorFlow on Raspberry Pi 3
https://github.com/samjabrahams/tensorflow-on-raspberry-pi
Compiling and running TensorFlow correctly on Raspberry Pi
Classification on time series
https://github.com/guillaume-chevalier/LSTM-Human-Activity-Recognition
Using LSTM in TensorFlow to classify mobile sensor data with recurrent neural networks
2. Models/Projects
Show, Attend and Tell
https://github.com/yunjey/show_attend_and_tell
An image caption generator based on attention mechanism (the attention mechanism is one of the cutting-edge hotspots in deep learning, allowing focus on different parts of the input to provide a series of understanding)
Neural Style
https://github.com/cysmith/neural-style-tf
Implementation of Neural Style (Neural Style is an algorithm that allows machines to mimic the painting style of existing artworks to recreate an image)
Pretty Tensor
https://github.com/google/prettytensor
PrettyTensor provides a high-level builder API
Neural Style
https://github.com/anishathalye/neural-style
Implementation of Neural Style
TensorFlow White Paper Notes
https://github.com/samjabrahams/tensorflow-white-paper-notes
Annotated notes and summaries of the TensorFlow white paper, along with SVG graphics and document links
NeuralArt
https://github.com/ckmarkoh/neuralart_tensorflow
Implementation of the artistic style neural algorithm
Generative Handwriting Demo using TensorFlow
https://github.com/hardmaru/write-rnn-tensorflow
Attempting to implement the random handwriting generation part from Alex Graves’ paper
Neural Turing Machine in TensorFlow
https://github.com/carpedm20/NTM-tensorflow
Implementation of the Neural Turing Machine in TensorFlow
GoogleNet Convolutional Neural Network Groups Movie Scenes By Setting
https://github.com/agermanidis/thingscoop
Searching, filtering, and describing videos based on objects, locations, and other content displayed
Chatbot
https://github.com/Conchylicultor/DeepQA
Implementation of a