How davidit17/ml works
High-Level Organization
• Divide the projects into two main categories:
a. Classification Projects (Blood Donation, Water Pump, Heart Disease, Insurance Accidents)
b. Regression Projects (Future Sales, Digital Turbine's Auction)
Project Components Structure
For each project, create a standardized representation with these elements:
a. Data Input Layer
• Data source/CSV files
• Data preprocessing steps
b. Model Layer
• Algorithms by project
• Rectangles for traditional ML (KNN, LogReg)
• Hexagons for ensemble methods (XGBoost)
• Clouds for neural networks (TensorFlow)
c. Output Layer
• Prediction results
• Model evaluation metrics
Specific Implementation Details
Blood Donation Project:
• [Input Data] → [Preprocessing] → [Models]
• KNN
• LogReg
• XGBoost
• TensorFlow
Similar structures for other projects, showing their specific algorithms.
Additional Elements
• Version information for major frameworks (TensorFlow 2.0)
• Evaluation metrics placement
Technical Details
• Framework versions (TensorFlow 2.0)
• File formats (CSV, notebook files)
• Model output formats
Metadata Section
• A small section showing common tools across projects
• Shared libraries and dependencies
• Common evaluation metrics
• Input data sources
• Preprocessing steps
• Model architecture
• Evaluation methods
• Output format
Remember to:
• Use consistent notation throughout
• Provide clear labels for all components
• Relationships between different components clearly
This structure will help other developers understand both the overall organization of the ML projects collection and the specific implementation details of each project.