掌握神经风格迁移 - Tensorflow、Keras 和 Python
上次更新时间：2024-11-20
课程售价: 2.9 元
联系右侧微信客服充值或购买课程
课程内容
- 1 - Introduction (免费)
- 2 - What is Neural Style Transfer (免费)
- 3 - About this Project
- 4 - Why Should we Learn
- 5 - Applications
- 6 - Why Keras and Python
- 7 - Why Google Colab
- 8 - Setup the Working Directory
- 9 - Contents in Directory
- 10 - Activate GPU
- 11 - Checking the availability and usage of GPUs
- 12 - Mount Google Drive to Google Colab
- 13 - Necessary library imports
- 14 - Setting the directory path
- 15 - Displaying the base image and the style reference
- 16 - Defining the desired dimensions
- 17 - Preprocesses an image
- 18 - Convert the generated image back to its original format
- 19 - Calculate the Gram matrix
- 20 - Calculates the style loss
- 21 - Calculates the content loss
- 22 - Calculates the total variation loss
- 23 - Loading the VGG19
- 24 - Creating a dictionary
- 25 - Building a feature extraction model
- 26 - Define the names of the style layers and the content layer
- 27 - Set the weights
- 28 - Calculates the total loss
- 29 - Computes the loss and gradients
- 30 - Set up the optimizer
- 31 - Preprocess the base image style reference image and combination image
- 32 - Perform the style transfer optimization loop
- 33 - Save and display the final generated image
课程内容
33个讲座
- 1 - Introduction (免费)
- 2 - What is Neural Style Transfer (免费)
- 3 - About this Project
- 4 - Why Should we Learn
- 5 - Applications
- 6 - Why Keras and Python
- 7 - Why Google Colab
- 8 - Setup the Working Directory
- 9 - Contents in Directory
- 10 - Activate GPU
- 11 - Checking the availability and usage of GPUs
- 12 - Mount Google Drive to Google Colab
- 13 - Necessary library imports
- 14 - Setting the directory path
- 15 - Displaying the base image and the style reference
- 16 - Defining the desired dimensions
- 17 - Preprocesses an image
- 18 - Convert the generated image back to its original format
- 19 - Calculate the Gram matrix
- 20 - Calculates the style loss
- 21 - Calculates the content loss
- 22 - Calculates the total variation loss
- 23 - Loading the VGG19
- 24 - Creating a dictionary
- 25 - Building a feature extraction model
- 26 - Define the names of the style layers and the content layer
- 27 - Set the weights
- 28 - Calculates the total loss
- 29 - Computes the loss and gradients
- 30 - Set up the optimizer
- 31 - Preprocess the base image style reference image and combination image
- 32 - Perform the style transfer optimization loop
- 33 - Save and display the final generated image