Combat: Google engineer Qinshou Tensorflow2.0- introductory to advanced

•  

• Chapter 1 Tensorflow Introduction and environmental structures

  Getting Started section of this course, what is a brief introduction tensorflow, detailing the history of architecture and powerful features Tensorflow version of changes and tensorflow. And between Tensorflow1.0, pytorch, Tensorflow2.0 to do a comparison. Finally, the actual environment configuration explained in the Google cloud and AWS two platforms.

  •  1-1 Course Guidance Look
  •  What is 1-2 Tensorflow
  •  1-3 Tensorflow version changes and tf1.0 architecture
  •  1-4 Tensorflow2.0 architecture Look
  •  1-5 Tensorflow & pytorch Comparative Look
  •  1-6 Tensorflow environment configuration
  •  1-7 Google_cloud GPU-free environment to build
  •  1-8 Google_cloud_ remote configuration jupyter_notebook
  •  1-9 Google_cloud_gpu_tensorflow配置
  •  1-10 Google_cloud_gpu_tensorflow mirrored configuration
  •  1-11 AWS cloud platform environment configuration

• Chapter 2 Tensorflow keras combat

  Basics section of this course, detailing the theoretical knowledge of how to use tf.keras be built models and a lot of depth learning. Theoretical knowledge, including classification, regression, loss of function, neural networks, activation function, dropout, batch normalization, deep neural networks, Wide & Deep model, dense features, sparse features, ultra-parameter search, and image classification, prices forecast implementation on. ...

  •  2-1 tfkeras Profile
  •  2-2 classification and regression and objective function
  •  2-3 classification model of actual data read and display
  •  2-4 classification model of the actual model building
  •  2-5 real data classification model of normalization
  •  2-6 combat callback function
  •  2-7 combat regression model
  •  2-8 to explain the neural network
  •  2-9 real depth neural network
  •  2-10 actual batch normalization, activation function, dropout
  •  2-11 wide_deep model
  •  2-12 API function to achieve wide & deep model
  •  2-13 subclass API to achieve wide & deep model
  •  2-14 wide & deep multi-input and multi-output model of the actual
  •  2-15 hyperparameter search
  •  2-16 manual search parameters to achieve ultra-practical
  •  2-17 sklearn actual package model keras
  •  2-18 combat sklearn hyperparameter search

• Chapter 3 Tensorflow basis using the API

•