Source | Yuedong Intelligent (Public ID: aibbtcom)
This article will summarize the TensorFlow pure dry goods learning resources for everyone, which is very suitable for novice learning. It is recommended that you collect them.
▌One , TensorFlow tutorial resources
1) TensorFlow tutorials and code samples for beginners:
https://github.com/aymericdamien/TensorFlow-Examples
This tutorial not only provides some classic data sets , but also starts from the simplest implementation of "Hello World", to the classic algorithms of machine learning , and then to the commonly used models of neural networks. It will take you step by step from entry to mastery. The best tutorials for academics to learn TensorFlow.
2) From TensorFlow basics to interesting project applications:
https://github.com/pkmital/tensorflow_tutorials
It is also a tutorial suitable for novices, from installation to actual project combat, teaching you to build a neural network of your own.
3) TensorFlow tutorial run with Jupyter Notebook:
https://github.com/sjchoi86/Tensorflow-101
4) TensorFlow Chinese Community
http://www.aibbt.com/a/tensorflow/
This tutorial is a TensorFlow tutorial based on the Jupyter Notebook development environment. Jupyter Notebook is a very useful interactive development tool that not only supports more than 40 programming languages, but also can run code in real time, share documents, data visualization, support markdown, etc. It is suitable for many fields such as machine learning, statistical modeling data processing, feature extraction, etc.
4) Build your first TensorFlow Android app:
https://omid.al/posts/2017-02-20-Tutorial-Build-Your-First-Tensorflow-Android-App.html
This tutorial helps you bring a tensorflow model to an Android application from scratch.
5) TensorFlow code exercise:
https://github.com/terryum/TensorFlow_Exercises
An easy-to-hard TensorFlow code practice manual. It is very suitable for small partners learning TensorFlow.
Next, I will recommend some good TensorFlow video tutorials for you:
▌Second , TensorFlow video resources
1) TF Girls training guide:
https://www.youtube.com/watchv=TrWqRMJZU8A&list=PLwY2GJhAPWRcZxxVFpNhhfivuW0kX15yG&index=2
A TensorFlow open video course from scratch. The course is basic and introductory, but the knowledge points are very detailed.
2) Alchemy into gold TensorFlow public class:
https://www.youtube.com/watchv=eAtGqz8ytOI&list=PLjSwXXbVlK6IHzhLOMpwHHLjYmINRstrk
Very good course, I recommend it to everyone.
3) Of course, the deep learning course of Li Hongyi's tutorial of National Taiwan University is also worth recommending to everyone:
https://www.bilibili.com/video/av9770302/
4) Friends with good English, I also recommend some English courses of foreign Daniels for everyone:
https://www.youtube.com/watch?v=vq2nnJ4g6N0;
http://bit.ly/1OX8s8Y;
https://www.youtube.com/watch?v=GZBIPwdGtkk&t=125s
5) With so many courses introduced, how can the courses of Stanford University's TensorFlow series be omitted! ! !
Not much to say, go directly to the link:
https://www.youtube.com/watch?v=g-EvyKpZjmQ&index=1&list=PLIDllPt3EQZoS8gCP3cw273Cq9puuPLTg
Course Homepage:
http://web.stanford.edu/class/cs20si/index.html
The download address of all ppts and notes of the course:
https://pan.baidu.com/s/1o8uOQpW
Course-related actual github address:
chiphuyen / tf-stanford-tutorials
6) Finally, how can we forget the video tutorial published by Google Dad on the TensorFlow official website. It is still a very good set of courses for the beginners of TensorFlow, which will help you get started quickly:
https://developers.google.cn/machine-learning/crash-course/
Well, through the resource documents and video tutorials above, you already have a solid foundation for TensorFlow. Should you do some high-level practical projects to improve yourself? Therefore, I will recommend some practical project resources for you.
▌3 . TensorFlow project resources
1) A case that implements the random handwriting generation of Alex Graves's paper:
https://github.com/hardmaru/write-rnn-tensorflow
2) TensorFlow-based Generative Adversarial Text-to-Image Synthesis:
https://github.com/zsdonghao/text-to-image
As shown in the figure below, this project is a DC GAN model based on TensorFlow, which teaches you step by step from confrontational text generation to image synthesis.
3) Attention-based image caption generator:
https://github.com/yunjey/show-attend-and-tell
This model introduces an attention-based image caption generator. It can shift its attention to the relevant part of the image, generating each word simultaneously.
4) Neural network colorize grayscale image:
https://github.com/pavelgonchar/colornet
A very interesting and widely used project, using neural networks to color grayscale images.
5) Simple embedded text classifier based on FastText in Facebook :
https://github.com/apcode/tensorflow_fasttext
This project was born out of the idea of FastText in Facebook and implemented in TensorFlow. FastText is a fast text classifier that provides simple and efficient methods for text classification and representation learning.
6) Implement " Convolutional Neural Network Based on Sentence Classification" with TensorFlow :
https://github.com/dennybritz/cnn-text-classification-tf
7) Train a TensorFlow neural network using the OpenStreetMap function and satellite imagery:
https://github.com/jtoy/awesome-tensorflow
The project is to classify features in satellite images by training a neural network using OpenStreetMap (OSM) data.
8) Implement YOLO with Tenflow: "Real-time Object Detection" and support a small project https://github.com/thtrieu/darkflow that runs on mobile devices in real-time, the best benefit for researchers in the field of computer vision .