Coursera 笔记

DeepLearning.AI

Custom Models, Layers, and Loss Functions with TensorFlow

使用 TensorFlow 自定义模型、层与损失函数

0 notes · 0/58 条官方条目有笔记 · 在学

Coursera ↗

Module 01

Functional APIs

0 / 18 篇
  1. 01A conversation with Andrew Ng: Overview of the specialization
  2. 02A conversation with Andrew Ng: Overview of course 1
  3. 03Welcome to the course
  4. 04Introduction to the Functional APIs
  5. 05Join the DeepLearning.AI Forum to ask questions, get support, or share amazing ideas!
  6. 06Declaring and stacking layers
  7. 07Branching models
  8. 08Learn more about the Inception Model Architecture
  9. 09Creating a Multi-Output model
  10. 10Energy efficiency dataset
  11. 11Multi-Output code walkthrough
  12. 12Siamese network: a Multiple-Input model
  13. 13References about the Siamese network
  14. 14Coding a Multi-Input Siamese network
  15. 15Reference "The distance between two vectors"
  16. 16Siamese network code walkthrough
  17. 17Lecture Notes Week 1
  18. 18(Optional) Downloading your Notebook and Refreshing your Workspace
Module 02

Custom Loss Functions

0 / 12 篇
  1. 01Welcome to Week 2
  2. 02Creating a custom loss function
  3. 03Huber Loss reference
  4. 04Coding the Huber Loss function
  5. 05Huber Loss code walkthrough
  6. 06Adding hyperparameters to custom loss functions
  7. 07Turning loss functions into classes
  8. 08Huber Object Loss code walkthrough
  9. 09Contrastive Loss
  10. 10Reference: Dimensionality reduction by Learning an Invariant Mapping
  11. 11Coding Contrastive Loss
  12. 12Lecture Notes Week 2
Module 03

Custom Layers

0 / 11 篇
  1. 01Intro custom layers
  2. 02Introduction to Lambda Layers
  3. 03Custom Functions from Lambda Layers
  4. 04Exploring custom Relu with Lambda Layers
  5. 05Architecture of a Custom Layer
  6. 06Coding your own custom Dense Layer
  7. 07Training a neural network with your Custom Layer
  8. 08Custom Layer code walkthrough
  9. 09Activating your Custom Layer
  10. 10Custom Layer with activation code walkthrough
  11. 11Lecture Notes Week 3
Module 04

Custom Models

0 / 10 篇
  1. 01Intro to custom models
  2. 02Complex architectures with the Functional API
  3. 03Coding a Wide and Deep model
  4. 04Using the Model class to simplify architectures
  5. 05Understanding Residual networks
  6. 06Residual networks lectures (optional)
  7. 07Coding a Residual network with the Model class
  8. 08ResNet code walkthrough
  9. 09Lecture Notes Week 4
  10. 10[IMPORTANT] Reminder about end of access to Lab Notebooks
Module 05

Bonus Content - Callbacks

0 / 7 篇
  1. 01Built-in Callbacks
  2. 02TensorBoard visualization toolkit
  3. 03Custom Callbacks
  4. 04Custom Callbacks code walkthrough
  5. 05Lecture Notes Week 5
  6. 06References
  7. 07Acknowledgments