Introduction to TensorFlow
In this course, you will learn how to create machine learning models in TensorFlow which is the tool we will use to write machine learning programs. You’ll learn how to use the TensorFlow libraries to solve numerical problems. When writing programs, you often want to know about common mistakes that you might run into, and how to fix common errors. Then, we’ll look at the Estimator API, which provides the highest level abstraction within TensorFlow for training, evaluating and serving machine learning models. You will learn how to Use tf_estimator to create, train, and evaluate an ML model. Finally, you’ll learn how to execute TensorFlow models on Cloud AI Platform, Google-managed infrastructure to run TensorFlow. You will learn how to Train, deploy, and productionalize ML models at scale with Cloud AI Platform.
What you'll learn
In this course, you will learn how to create machine learning models in TensorFlow which is the tool we will use to write machine learning programs. You’ll learn how to use the TensorFlow libraries to solve numerical problems. When writing programs, you often want to know about common mistakes that you might run into, and how to fix common errors. Then, we’ll look at the Estimator API, which provides the highest level abstraction within TensorFlow for training, evaluating and serving machine learning models. You will learn how to Use tf_estimator to create, train, and evaluate an ML model. Finally, you’ll learn how to execute TensorFlow models on Cloud AI Platform, Google-managed infrastructure to run TensorFlow. You will learn how to Train, deploy, and productionalize ML models at scale with Cloud AI Platform.
Table of contents
- Module 2 Slides 0m
- Introduction 2m
- What is TensorFlow 3m
- Benefits of a DAG 6m
- TensorFlow API Hierarchy 4m
- Lazy Evaluation 5m
- Graph and Session 4m
- Evaluating a Tensor 2m
- Visualizing a graph 3m
- Tensors 6m
- Variables 7m
- Lab Intro:Writing low-level TensorFlow programs 0m
- Lab: Writing low-level TensorFlow programs 0m
- Lab Solution 8m
- Introduction to Debugging TensorFlow Programs 6m
- Challenge: Shape problems 4m
- Fixing shape problems 3m
- Data type problems 2m
- Debugging full programs 4m
- Demo Intro:Debugging full programs 0m
- Demo:Debugging full programs 4m
- Module 3 Slides 0m
- Introduction 1m
- Estimator API 3m
- Pre-made Estimators 5m
- Demo:Housing Price Model 2m
- Checkpointing 2m
- Training on in-memory datasets 3m
- Lab Intro:Estimator API 1m
- Lab: Implementing a AI model in TensorFlow using Estimator API 0m
- Lab Solution:Estimator API 11m
- Train on large datasets with Dataset API 8m
- Lab Intro:Scaling up TensorFlow ingest using batching 1m
- Lab: Scaling up TensorFlow ingest using batching 0m
- Lab Solution:Scaling up TensorFlow ingest using batching 5m
- Big jobs, Distributed training 6m
- Monitoring with TensorBoard 4m
- Demo:TensorBoard UI 0m
- Serving Input Function L15a 6m
- Serving Input Function L15b 1m
- Lab Intro:Creating a distributed training TensorFlow model with Estimator API 1m
- Lab: Creating a distributed training TensorFlow model with Estimator API 0m
- Lab Solution:Creating a distributed training TensorFlow model with Estimator API 7m
- Summary 1m
