Practical case of TensorFlow 2.0/TF2.0 introductory course


Explain machine learning, neural networks and in-depth learning in the most vernacular language
The example is implemented based on TensorFlow 1.4 and TensorFlow 2.0

Related links

  • Machine Learning Written Examination Interview Question, Github
  • TensorFlow 2.0 Chinese Document, Github
  • TensorFlow 2.0 Image Recognition & Enhanced Learning Practice, Github

OpenAI gym

  • TensorFlow 2.0 (9) – Enhanced Learning 70 Lines of Code Practical Policy Gradient

    • Github – gym/CartPole-v0-policy-gradient
    • This paper introduces the Policy Gradient algorithm to play CartPole-v0.
  • TensorFlow 2.0 (8) – Enhanced Learning DQN Playing gym Mountain Car

    • Github – gym/MountainCar-v0-dqn
    • This paper introduces DQN (Deep Q-Learning) to play MountainCar-v0 game.
    • Q-Table is replaced by a neural network.
  • TensorFlow 2.0 (7) – Enhanced Learning Q-Learning Plays OpenAI gym

    • Github – gym/MountainCar-v0-q-learning
    • This paper introduces the use of Q-Learning (creating Q-Table) to play MountainCar-v0 game.
    • The continuous state is discretized.
  • TensorFlow 2.0 (6) – Supervised Learning Play OpenAI gym game

    • Github – gym/CartPole-v0-nn
    • This paper introduces the use of pure supervised learning (neural network) to play CartPole-v0 game.
    • Using TensorFlow 2.0


  • TensorFlow 2.0 (V) – MNIST Handwritten Digital Recognition (CNN Convolutional Neural Network)

    • Github – v4_cnn
    • This paper introduces how to build CNN network, the accuracy rate is 0.99.
    • Using TensorFlow 2.0
  • TensorFlow Introduction (IV) – MNIST Handwritten Number Recognition (Making h5py Training Set)

    • Github – make_data_set
    • This paper introduces how to use numpy to make NPY format data sets.
    • This paper introduces how to use h5py to make data sets in HDF5 format.
  • Introduction to TensorFlow (III) – MNIST Handwritten Number Recognition (Visual Training)

    • Github – mnist/v3
    • This paper introduces the simple usage of tensorboard, including scalar graph, histogram and network structure chart.
  • Introduction to TensorFlow (II) – MNIST Handwritten Number Recognition (Model Preservation Loading)

    • Github – mnist/v2
    • This paper introduces how to save the trained model in TensorFlow.
    • This paper introduces how to continue training from a certain model.
    • This paper introduces how to load and use the model, and how to recognize the real image.
  • Introduction to TensorFlow (I) – MNIST Handwritten Digital Recognition (Network Construction)

    • Github – mnist/v1
    • This blog introduces the use of TensorFlow to build the simplest neural network.
    • Including input and output, exclusive thermal coding and loss function, as well as the verification of accuracy.

Recommended Today

Implementation of PHP Facades

Example <?php class RealRoute{ public function get(){ Echo’Get me’; } } class Facade{ public static $resolvedInstance; public static $app; public static function __callStatic($method,$args){ $instance = static::getFacadeRoot(); if(!$instance){ throw new RuntimeException(‘A facade root has not been set.’); } return $instance->$method(…$args); } // Get the Facade root object public static function getFacadeRoot() { return static::resolveFacadeInstance(static::getFacadeAccessor()); } protected […]