GCP for Data Scientists
summary
A pipeline shows how a Raspberry Pi streams audio to Google Cloud: recordings are uploaded to Cloud Storage, a Cloud Function splits each WAV into five segments, a Python/Anaconda feature extractor runs, and a model deployed on Cloud ML Engine scores each segment. The five segment predictions are combined by a 3-of-5 majority vote and the decision is sent back to the Pi; the model was trained locally and deployed, with end-to-end inference under 5 seconds in live tests.
AI-written summary · about
chapters (12)
- 0:00Presentation by Giulia
- 0:36Why GCP?
- 2:18What is the Cloud?
- 3:10Why the Cloud?
- 4:15GCP and ML
- 5:09Machine Learning
- 5:59Baby Detection
- 8:15Prediction Phases
- 9:55The Final Prediction
- 11:07Some Code Snippets
- 12:21Reading
- 12:38Conclusion
related talks
34mGCP for Data Scientists
53mCloud Machine Learning APIs for fun and profit with Harry Potter
42mBringing Clinical Data to GCP for Automated AI Discoveries
37mPracticing Data Science in the cloud
27mUsing GenAI with Audio Data for Driving Better Customer Experience
23mBuilding Scalable, Observable MLOps Systems on Google Cloud
19mData science for real problems
59mMachine Learning and Big Data Magic with Node.js and Google Cloud