In this video, learn about the Cloud ML models and how they are used.
- [Instructor] Machine learning in Cloud ML…is built on models.…A model is a logical container of solutions…for a specific problem.…For example, we have a problem of predicting spam emails…for this problem, we may build a predictor model.…This predictor model is stored in GCP…as a Cloud ML model.…All models are members of a specific GCP project.…The model names are unique within a GCP project.…
A model can contain multiple versions of the solution.…As you experiment with machine learning,…you can build several versions of the same model…with varying algorithms or settings.…All of them can be deployed and used…within a single Cloud ML model as versions.…This allows for comparing model performance…and doing A/B testing.…
Author
Kumaran Ponnambalam
Released
11/7/2018- Evaluating the machine learning tools in GCP
- Understanding the predictive analytics process
- Building models
- Training models with jobs
- Building and running predictions
- Best practices for cost control, testing, and performance monitoring
Skill Level Intermediate
Duration
Views
-
Introduction
-
1. ML Options in GCP
-
Cloud Dataproc56s
-
Cloud ML Engine1m 37s
-
Cloud Natural Language1m 20s
-
Cloud Translation1m 17s
-
Cloud Vision1m 18s
-
Cloud Video Intelligence1m 2s
-
Cloud Dialogflow1m 14s
-
-
2. Cloud ML Basics
-
Models56s
-
Model versions41s
-
Jobs56s
-
Predictive analytics process1m 55s
-
-
3. Model Building with Cloud ML
-
Understanding input data1m 30s
-
Build and test model locally1m 53s
-
Modify code to work with GCP1m 42s
-
Creating a training package1m 21s
-
Training using jobs3m 49s
-
-
4. Predictions in Cloud ML
-
Creating a model version2m 10s
-
Running a prediction1m 37s
-
5. Cloud ML Best Practices
-
Cost control1m 17s
-
Local testing1m 12s
-
Performance monitoring1m 35s
-
-
Conclusion
-
Next steps41s
-
Take notes with your new membership!
Type in the entry box, then click Enter to save your note.
Press on any video thumbnail to jump immediately to the timecode shown.
Notes are saved with you account but can also be exported as plain text, MS Word, PDF, Google Doc, or Evernote.
Share this video
Embed this video
Video: Models