MLOps

1.1. Introduction

03:50

2.1. ML Project Life Cycle

09:41

3.1. ML life Cycle: Case Study

15:49

4.1. Deployment

12:52

5.1. Common Deployment Patterns

08:44

6.1. Monitoring

10:29

7.1. Pipeline Monitoring

09:25

8.1. Modeling

06:36

9.1. Low Average Test Error isn't Good Enough

06:35

10.1. Establish a Baseline

13:33

11.1. Performance Auditing

21:01

12.1. Data-centric AI

21:11

13.1. Data Definition

13:20

14.1. Label Consistency

11:16

15.1. HLP

10:54

16.1. Data Pipeline

17:19

17.1. Scoping

19:29

1.2. Data Lifecycle

15:16

2.2. Data

15:31

3.2. Responsible Data

20:53

4.2. Labelling

08:30

5.2. Data Issues

10:17

6.2. Feature Engineering

09:41

7.2. Preprocessing in Production

13:35

8.2. TF Transform

12:37

9.2. Feature Selection Example

09:18

10.2. ML Metadata

09:11

11.2. Evolving Data

13:15

12.2. Advanced Labeling

10:40

1.3. Neural Architecture Search

11:14

2.3. Auto ML

11:40

3.3. Resource Management

12:02

4.3. Quantization and Pruning

19:30

5.3. High Performance Modeling

17:08

6.3. Model analysis

09:12

7.3. Robustness

15:31

8.3. Remediation and Monitoring

15:28

9.3. Interpretability

22:32

1.4. Model Serving

13:50

2.4. Serving Systems

10:53

3.4. Serving Patterns

14:26

4.4. Online and Batch Inference

12:52

5.4. Experiment Tracking

20:01

6.4. Mlops level 0, 1 and 2

13:57

7.4. CI/CD

09:47

8.4. Monitoring and Privacy

23:05

1.5. FastAPI and a Simple Deep Model

15:02

2.5. TensorFlow Data Validation

10:39

3.5. Preprocess Data with TensorFlow Transform

08:40

4.5. Simple TFX Pipeline

13:57

5.5. Recursive Feature Elimination

06:42

6.5. Active Learning

14:49

7.5. Keras Tuner

14:07

8.5. Post-Training Quantization

09:43

9.5. Quantization Aware Training

07:12

10.5. Pruning

09:04

11.5. Distributed Training: One Device Strategy

06:23

12.5. Knowledge Distillation

08:28