Enumerate the stages of predictive analytics and how they are achieved on Cloud ML in this video.
- In this video, I will review the various steps…in the predictive analytics process…and illustrate how this process is achieved in Cloud ML.…Predictive analytics starts with setting a goal…for solving a problem.…Usually, it involves predicting the behavior of an element,…or outcome of an activity.…First, data needs to be acquired from its sources.…It then goes through processing, which might involve…cleansing, transformations, enhancements, and aggregations.…
Then, a training data set is created…on which the model is built.…The model is tested against the test data set.…Once the model is confirmed to be good,…it is then used for predicting future behavior or outcomes.…How does the predictive analytics process work in Cloud ML?…The work of collecting data, processing and building…the model are usually done offline.…GCP components like Pub/Sub, BigQuery, Cloud DataFlow, and…Cloud Storage may also be used for this process.…
Model building can also be done as Cloud ML jobs,…but they can be expensive.…Next, a Cloud ML model and its versions are set up…
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: Predictive analytics process