计算机视觉机器学习实践（影印版）
出版时间：2022年06月
页数：463
“本书全面介绍了深度计算机视觉的最新技术，以及在 Keras中构建端到端生产系统以解决实际问题的最佳实践。”
一Francois Chollet深度学习研究员以及Keras创造者
本书向你展示了如何使用机器学习（ML）模型从图像中提取信息。ML工程师和数据科学家将学习行之有效的ML技术来解决各种图像问题，包括分类、目标检测、自动编码器、图像生成、计数和添加字幕。本书很好地介绍了端到端的深度学习:数据集创建、数据预处理、模型设计、模型训练、评估、部署和可解释性。
Google工程师Valliappa Lakshmanan、Martin Gorner、Ryan Gillard为你展示了如何开发精准且可解释的计算机视觉ML模型，并以灵活且可维护的方式使用健壮的ML架构将其投入大规模生产。你将学习如何使用TensorFlow和Keras编写的模型进行设计、训练、评估和预测。
你会学到:
● 为计算机视觉任务设计ML架构
● 选择适合任务的模型(比如ResNet、SqueezeNet或EfficientNet)
● 创建端到端的ML管道来训练、评估、部署、解释你的模型
● 对图像进行预处理，以增强数据并支持可学习性
● 结合可解释性和可靠的AI最佳实践
● 将图像模型部署为Web服务或边缘设备
● 监测和管理ML模型
● 使用协作过滤设计出一套仿Netflix的推荐系统
一Francois Chollet深度学习研究员以及Keras创造者
本书向你展示了如何使用机器学习（ML）模型从图像中提取信息。ML工程师和数据科学家将学习行之有效的ML技术来解决各种图像问题，包括分类、目标检测、自动编码器、图像生成、计数和添加字幕。本书很好地介绍了端到端的深度学习:数据集创建、数据预处理、模型设计、模型训练、评估、部署和可解释性。
Google工程师Valliappa Lakshmanan、Martin Gorner、Ryan Gillard为你展示了如何开发精准且可解释的计算机视觉ML模型，并以灵活且可维护的方式使用健壮的ML架构将其投入大规模生产。你将学习如何使用TensorFlow和Keras编写的模型进行设计、训练、评估和预测。
你会学到:
● 为计算机视觉任务设计ML架构
● 选择适合任务的模型(比如ResNet、SqueezeNet或EfficientNet)
● 创建端到端的ML管道来训练、评估、部署、解释你的模型
● 对图像进行预处理，以增强数据并支持可学习性
● 结合可解释性和可靠的AI最佳实践
● 将图像模型部署为Web服务或边缘设备
● 监测和管理ML模型
● 使用协作过滤设计出一套仿Netflix的推荐系统
- Preface
- 1. Machine Learning for Computer Vision
- Machine Learning
- Deep Learning Use Cases
- Summary
- 2. ML Models for Vision
- A Dataset for Machine Perception
- A Linear Model Using Keras
- A Neural Network Using Keras
- Summary
- Glossary
- 3. Image Vision
- Pretrained Embeddings
- Convolutional Networks
- The Quest for Depth
- Modular Architectures
- Neural Architecture Search Designs
- Beyond Convolution: The Transformer Architecture
- Choosing a Model
- Summary
- 4. Object Detection and Image Segmentation
- Object Detection
- Segmentation
- Summary
- 5. Creating Vision Datasets
- Collecting Images
- Data Types
- Manual Labeling
- Labeling at Scale
- Automated Labeling
- Bias
- Creating a Dataset
- Summary
- 6. Preprocessing
- Reasons for Preprocessing
- Size and Resolution
- Training-Serving Skew
- Data Augmentation
- Forming Input Images
- Summary
- 7. Training Pipeline
- Efficient Ingestion
- Saving Model State
- Distribution Strategy
- Serverless ML
- Summary
- 8. Model Quality and Continuous Evaluation
- Monitoring
- Model Quality Metrics
- Quality Evaluation
- Summary
- 9. Model Predictions
- Making Predictions
- Online Prediction
- Batch and Stream Prediction
- Edge ML
- Summary
- 10. Trends in Production ML
- Machine Learning Pipelines
- Explainability
- No-Code Computer Vision
- Summary
- 11. Advanced Vision Problems
- Object Measurement
- Counting
- Pose Estimation
- Image Search
- Summary
- 12. Image and Text Generation
- Image Understanding
- Image Generation
- Image Captioning
- Summary
- Afterword
- Index
书名：计算机视觉机器学习实践（影印版）
国内出版社：东南大学出版社
出版时间：2022年06月
页数：463
书号：978-7-5641-9976-0
原版书书名：Practical Machine Learning for Computer Vision
原版书出版商：O'Reilly Media
The bird on the cover of Practical Machine Learning for Computer Vision is an emerald toucanet (Aulacorhynchus prasinus), the smallest species of toucan. Central and South America have large populations from the cloud forests of Costa Rica to Venezuela.
Vibrant green feathers camouflage emerald toucanets in the tropics. Adults typically measure 12–13 inches long, weigh just over 5 ounces, and live 10–11 years in the wild. Their beaks are colorful: yellow on top, a white outline, and red or black on the bottom. They eat fruit and insects, as well as small lizards and the eggs and young of other birds. Groups of about eight will hunt and forage together. Emerald toucanets build their nests by enlarging the nests of smaller birds. The male and female trade off shifts in the nest, incubating, feeding, and cleaning their chicks.
Deforestation has driven emerald toucanets into shade coffee farms. Overall, their population is decreasing. Many of the animals on O’Reilly’s covers are endangered; all of them are important to the world.
Vibrant green feathers camouflage emerald toucanets in the tropics. Adults typically measure 12–13 inches long, weigh just over 5 ounces, and live 10–11 years in the wild. Their beaks are colorful: yellow on top, a white outline, and red or black on the bottom. They eat fruit and insects, as well as small lizards and the eggs and young of other birds. Groups of about eight will hunt and forage together. Emerald toucanets build their nests by enlarging the nests of smaller birds. The male and female trade off shifts in the nest, incubating, feeding, and cleaning their chicks.
Deforestation has driven emerald toucanets into shade coffee farms. Overall, their population is decreasing. Many of the animals on O’Reilly’s covers are endangered; all of them are important to the world.
购买选项
|
© 2026，奥莱利技术咨询(北京)有限公司。版权所有。 传真: +86-10-88097463 京ICP备05003502号 |
|