MLOps

1.1. Introduction

03

2.1. ML Project Life Cycle

09

3.1. ML life Cycle: Case Study

15

4.1. Deployment

12

5.1. Common Deployment Patterns

08

6.1. Monitoring

10

7.1. Pipeline Monitoring

09

8.1. Modeling

06

9.1. Low Average Test Error isn't Good Enough

06

10.1. Establish a Baseline

13

11.1. Performance Auditing

21

12.1. Data-centric AI

21

13.1. Data Definition

13

14.1. Label Consistency

11

15.1. HLP

10

16.1. Data Pipeline

17

17.1. Scoping

19

1.2. Data Lifecycle

15

2.2. Data

15

3.2. Responsible Data

20

4.2. Labelling

08

5.2. Data Issues

10

6.2. Feature Engineering

09

7.2. Preprocessing in Production

13

8.2. TF Transform

12

9.2. Feature Selection Example

09

10.2. ML Metadata

09

11.2. Evolving Data

13

12.2. Advanced Labeling

10

1.3. Neural Architecture Search

11

2.3. Auto ML

11

3.3. Resource Management

12

4.3. Quantization and Pruning

19

5.3. High Performance Modeling

17

6.3. Model analysis

09

7.3. Robustness

15

8.3. Remediation and Monitoring

15

9.3. Interpretability

22

1.4. Model Serving

13

2.4. Serving Systems

10

3.4. Serving Patterns

14

4.4. Online and Batch Inference

12

5.4. Experiment Tracking

20

6.4. Mlops level 0, 1 and 2

13

7.4. CI/CD

09

8.4. Monitoring and Privacy

23

1.5. FastAPI and a Simple Deep Model

15

2.5. TensorFlow Data Validation

10

3.5. Preprocess Data with TensorFlow Transform

08

4.5. Simple TFX Pipeline

13

5.5. Recursive Feature Elimination

06

6.5. Active Learning

14

7.5. Keras Tuner

14

8.5. Post-Training Quantization

09

9.5. Quantization Aware Training

07

10.5. Pruning

09

11.5. Distributed Training: One Device Strategy

06

12.5. Knowledge Distillation

08