Live Course Module: Microsoft Azure cloud for Data Science

Total Duration: 36 Hours (6 Weeks)

---

Week 1: Introduction & GCP Foundations (6 Hrs)

1. Session 1 (2 Hrs):

   • Introduction to GCP for Data Science

   • Key services (Compute Engine, Storage, BigQuery, AI/ML APIs)

   • Understanding GCP Console, SDK, and Cloud Shell

2. Session 2 (2 Hrs):

   • Setting up GCP Projects & IAM (Identity & Access Management)

   • Cloud Storage: Buckets, Data Import, Permissions

   • Hands-on: Uploading and organizing datasets

3. Session 3 (2 Hrs):

   • Networking basics in GCP (VPC, firewall rules)

   • Billing & cost optimization for data projects

   • Hands-on: Setting up billing alerts and quotas

---

Week 2: Data Engineering on GCP (6 Hrs)

4. Session 4 (2 Hrs):

   • Introduction to Data Engineering workflow in GCP

   • Cloud Storage → BigQuery pipeline

   • Hands-on: Loading CSV/JSON datasets into BigQuery

5. Session 5 (2 Hrs):

   • Dataflow & Pub/Sub for real-time pipelines

   • ETL basics in GCP

   • Hands-on: Stream data into BigQuery using Pub/Sub

6. Session 6 (2 Hrs):

   • Cloud Dataprep for data cleaning and transformation

   • Hands-on: Preparing datasets for ML models

---

Week 3: BigQuery for Data Science (6 Hrs)

7. Session 7 (2 Hrs):

   • Introduction to BigQuery

   • Writing SQL for analytics

   • Hands-on: Exploratory Data Analysis (EDA) in BigQuery

8. Session 8 (2 Hrs):

   • BigQuery ML: Basics of creating ML models in SQL

   • Regression & Classification models

   • Hands-on: Building a logistic regression model

9. Session 9 (2 Hrs):

   • Advanced BigQuery ML (time series, clustering)

   • Model evaluation & deployment in BigQuery

   • Hands-on: Predictive analytics project

---

Week 4: Machine Learning on GCP (6 Hrs)

10. Session 10 (2 Hrs):

    • Introduction to Vertex AI

    • AutoML for image/text/tabular data

    • Hands-on: Training a classification model with AutoML

11. Session 11 (2 Hrs):

    • Custom ML with TensorFlow & Scikit-learn on GCP

    • Using AI Platform Notebooks (JupyterLab on GCP)

    • Hands-on: Training ML model in Vertex AI Notebook

12. Session 12 (2 Hrs):

    • Hyperparameter tuning in Vertex AI

    • Model versioning & deployment as REST API

    • Hands-on: Deploy and test a prediction API

---

Week 5: Advanced AI & Analytics (6 Hrs)

13. Session 13 (2 Hrs):

    • Pre-trained APIs (Vision, NLP, Translation, Speech-to-Text)

    • Hands-on: Sentiment analysis using Cloud NLP API

14. Session 14 (2 Hrs):

    • Building recommendation systems in GCP

    • Vertex AI Pipelines for end-to-end ML lifecycle

    • Hands-on: Pipeline automation demo

15. Session 15 (2 Hrs):

    • BigQuery BI Engine for dashboards

    • Integration with Data Studio/Looker

    • Hands-on: Building interactive dashboards

---

Week 6: Capstone Project & Wrap-up (6 Hrs)

16. Session 16 (2 Hrs):

    • Capstone Project Overview (Data to AI Deployment)

    • Guidance on project planning & dataset selection

17. Session 17 (2 Hrs):

    • Team project work (real-world dataset on GCP)

    • Instructor guidance & troubleshooting

18. Session 18 (2 Hrs):

    • Project presentations & feedback

    • Course wrap-up, certifications, and next steps

---

✅ Final Outcome:
By end of the course, learners will:

• Build and deploy ML models using Vertex AI & BigQuery ML

• Perform data engineering workflows with Dataflow, Pub/Sub, Dataprep

• Create end-to-end pipelines from raw data → insights → deployed AI models

• Gain hands-on experience with real-world datasets on GCP