Módulo 1: End-to-End Machine Learning with TensorFlow on GCP
Temas:
In the first course of this specialization, we recap what was covered in the Machine Learning on Google Cloud Specialization. One of the best ways to review something is to work with the concepts and technologies that you have learned. So, this course is run like a workshop where you will carry out end-toend machine learning with TensorFlow on Google Cloud Platform. Here you will learn how to explore large datasets for features, create training and evaluation datasets, build models with the Estimator API in TensorFlow, train at scale and deploy those models into production with Google Cloud Platform machine learning tools.
New learners with ML background can also follow this course to learn how to do ML on GCP to fast track to the more advanced topics coming soon under the advanced specialization.
Módulo 2: Production ML Systems
Temas:
We’ll cover how to implement the various flavors of production ML systems—static, dynamic, and continuous training; static and dynamic inference; and batch and online processing. We’ll delve into TensorFlow abstraction levels and the various options for doing distributed training and how to write distributed training models with custom estimators.
- Compare static vs. dynamic training and inference
- Manage model dependencies
- Set up distributed training for fault tolerance, replication, and more
- Export models for portability
Módulo 3: Image Classification Models
Temas:
We will take a look at different strategies for building an image classifier using convolutional neural networks. We’ll improve the model’s accuracy with augmentation, feature extraction, and fine-tuning hyperparameters while avoiding overfitting our data.
- Classify images using deep learning
- Implement convolutional neural networks
- Improve the model by augmentation, batch normalization, etc.
- Leverage transfer learning
Objetivos:
Gain an overview of how ML is applied to image classification, including the evolving methods and challenges
Módulo 4: Sequence Models
Temas:
- Predict future values of a time-series
- Classify free form text
- Address time-series and text problems with recurrent neural networks
- Choose between RNNs/LSTMs and simpler models
- Train and reuse word embeddings in text problems
Objetivos:
This module is an introduction to sequence models and their applications, including an overview of sequence model architectures and how to handle inputs of variable length.
Módulo 5: Recommendation Models
Temas:
- Devise a content-based recommendation engine
- Implement a collaborative filtering recommendation engine
- Build a hybrid recommendation engine with user and content embeddings
Objetivos:
Apply your knowledge of classification models and embeddings to build a ML pipeline that functions as a recommendation engine.