Workflow Element Store

  1. Public Datasets
  2. Crowdsourcing
  3. Unstructured data (Images / Videos)
  4. Data Collaboration and Partnerships
  5. Data Pre-existing
  6. Data Generation
  7. WebScraping
  8. Unstructured data (Audio)
  9. Data Logging
  10. Structured Data (Tabular)
  11. Surveys and Questionnaires
  12. Mobile Applications or IoT Applications
  13. APIs and Data Feeds
  1. MySQL
  2. GCP BigQuery
  3. Informatica
  4. NoSQL DB
  5. S3
  6. Azure blob storage
  7. Oracle DB
  8. GCS
  9. AWS Redshift
  10. RDBMS
  11. MS SQL server
  12. Azure Data Warehouse
  13. PostgreSQL
  1. Handling Missing Data
  2. Domain-Specific Feature Engineering
  3. Time-Based Features
  4. Dealing with Outliers
  5. Dimensionality Reduction
  6. Feature Extraction from Images
  7. Feature Selection
  8. Handling Categorical Data
  9. Data Scaling and Normalization
  10. Handling Noisy Data
  11. Data Scaling and Normalization
  12. Auto-Preprocessing libraries
  13. Binning
  14. Handling Time-Series Data
  15. Polynomial Features
  16. Encoding Categorical Variables
  17. Textual Feature Extraction
  18. Handling Imbalanced Classes
  19. Interaction Features
  20. AutoEDA libraries
  21. Logarithmic Transform
  22. Dimensionality Reduction
  1. Ensemble Techniques
  2. Supervised Learning-Regression
  3. Unsupervised Learning
  4. Forecasting
  5. Time Series Anaysis
  6. Blackbox Techniques
  7. Supervised Learning-binary classification
  8. Supervised Learning-multiclass classification
  9. Data Partitioning
  10. Train-Test Split
  1. Data Augmentation
  2. Train-Test Split
  3. Regular Monitoring and Logging
  4. Ensemble Methods
  5. Cross-Validation
  6. Learning Rate Scheduling
  7. Batch Size Selection
  8. Gradient Clipping
  9. Data Partition-sequential
  10. Weight Initialization
  11. Regularization
  12. Hyperparameter Tuning
  13. Early Stopping
  14. Transfer Learning
  15. Batch Normalization
  1. Performance Visualization
  2. Data Partitioning
  3. Model Interpretability
  4. External Validation
  5. Regularization Techniques
  6. Evaluation Metrics
  7. Hyperparameter Tuning
  8. Train-Test Split
  9. Model Comparison
  10. Cross-Validation
  1. Data Drift Monitoring
  2. Model Registry
  3. Edge Deployment
  4. Documentation and API Documentation
  5. Alerting and Notification
  6. Web APIs - Flask, FastAPI, etc.
  7. Security Considerations
  8. Model Monitoring and Maintenance
  9. Prediction Logging
  10. Streamlit
  11. Model Health Monitoring
  12. Containerization
  13. A/B Testing
  14. Error Analysis
  15. Model Versioning
  16. Documentation and Reporting
  17. Monitoring and Logging
  18. Continuous Integration and Deployment (CI/CD)
  19. Cloud Deployment
  20. Model Drift
  21. Performance Metrics
  22. Model Retraining and Updating
  23. Concept Drift Detection
  24. Feedback Collection
  25. Serverless Computing
  26. Model Serialization
  27. Bias and Fairness Assessment
  1. End User Machine
  2. Mobile
ML Workflow - Architecture
  • Element belongs to model
  • Element not belongs to model

Feature Store
(Online / Offline)

Data Sources

Data Warehouse/ Data Lake

EDA, Data Pre Processing & Feature Engineering

Model Selection

Model Training & Hyper Parameter Tuning

Model Evaluation

Model Deployment

End User Device

Model Registry