Этот материал находится в платной подписке. Оформи премиум подписку и смотри или слушай The Real-World ML Tutorial, а также все другие курсы, прямо сейчас!
Премиум
  • Урок 1. 00:00:50
    Welcome
  • Урок 2. 00:00:57
    Development tools: VSCode, Python Poetry and Git/GitHub
  • Урок 3. 00:02:18
    Install VSCode and Python Poetry in your local machine
  • Урок 4. 00:02:19
    Create the Project Structure
  • Урок 5. 00:02:05
    Create local git repository
  • Урок 6. 00:01:55
    Create remote GitHub repository and connect it to the local one
  • Урок 7. 00:07:34
    Let's understand the business problem
  • Урок 8. 00:01:46
    3 Steps to go from raw data to training data
  • Урок 9. 00:01:39
    Our raw data source: the NYC taxi website
  • Урок 10. 00:09:22
    Step 1. Fetch raw data and validate it
  • Урок 11. 00:06:13
    Step 2. Transform raw validated data into time-series data
  • Урок 12. 00:02:44
    Plot the time-series data
  • Урок 13. 00:09:38
    Step 3. Transform time-series data into training data
  • Урок 14. 00:09:56
    Steps 1, 2 and 3. From raw data to training data + Code re-factoring!
  • Урок 15. 00:05:10
    Plot the training data
  • Урок 16. 00:09:32
    How do you build a Supervised Machine Learning model?
  • Урок 17. 00:01:47
    Split the dataset into training and test datasets
  • Урок 18. 00:06:28
    Baseline model 1
  • Урок 19. 00:02:28
    Baseline model 2
  • Урок 20. 00:02:40
    Baseline model 3
  • Урок 21. 00:05:27
    XGBoost model
  • Урок 22. 00:03:08
    LightGBM model
  • Урок 23. 00:11:40
    LightGBM + Feature engineering
  • Урок 24. 00:10:33
    LightGBM + Feature engineering + Hyper-parameter tuning
  • Урок 25. 00:03:56
    Batch-scoring ML service with a Feature Store
  • Урок 26. 00:01:18
    What is a Feature Store?
  • Урок 27. 00:01:42
    Create a Serverless Feature Store with Hopsworks
  • Урок 28. 00:08:52
    Backfill Feature Store with Historical Data
  • Урок 29. 00:05:22
    Build the Feature Pipeline
  • Урок 30. 00:04:47
    Automate the execution of the Feature Pipeline using a GitHub action
  • Урок 31. 00:09:15
    Build the Model Training Pipeline
  • Урок 32. 00:02:15
    ML Frontend app using Streamlit
  • Урок 33. 00:04:13
    Inference functions
  • Урок 34. 00:09:43
    Build the Streamlit app - Part 1: inference code
  • Урок 35. 00:05:41
    Build the Streamlit app - Part 2: build the UI
  • Урок 36. 00:04:54
    Deploy the Streamlit app to Streamlit Cloud
  • Урок 37. 00:02:35
    Our plan
  • Урок 38. 00:07:24
    Create an inference pipeline to generate and store predictions in the store
  • Урок 39. 00:02:56
    The new (and way simpler) frontend Streamlit app - frontend.py
  • Урок 40. 00:04:07
    Create a monitoring dashboard with Streamlit
  • Урок 41. 00:02:33
    Deploy the monitoring dashboard to Streamlit Cloud
  • Урок 42. 00:03:02
    Why model re-training?
  • Урок 43. 00:10:29
    Implementation
  • Урок 44. 00:05:11
    2024_07_02_Karthikeya
  • Урок 45. 00:25:20
    2024-07-09-Karthikeya