Skip to main content

Pipe Breaks and LOE COE Project Review

1. Project Overview​

Project Names​

  • Pipe Breaks: Original machine learning project for pipe break prediction
  • LOE COE: Likelihood-of-Event analysis implementation

Purpose​

The project consists of machine learning models focused on:

  • Predicting pipe breaks using historical data
  • Analyzing likelihood-of-event scenarios
  • Implementing transfer learning techniques

Key Components​

  1. Original Pipe Breaks Project
  2. LOE Application
  3. Transfer Learning Implementation

2. Repository Structure​

Main Repositories​

  • pipe-breaks: Original project repository
  • loe-coe-app_deven-fork: Fork of the LOE application
  • pipe-breaks-transfer-learning: Transfer learning implementation

Documentation Locations​

  • Local: ./internal/Pipe Breaks/
  • Remote:
    • https://github.com/gqc/pipe_breaks/tree/master/docs
    • Sphinx documentation in loe-coe-app (gh-pages branch)

3. Environment Details​

Development Machines​

  • HP Laptop
  • MSI

Python Environments​

Environment NamePython VersionKey Dependencies
pipe_breaks3.8.10scikit-learn=1.2.2
pipe-breaks3.9.16-
pipe_breaks_transfer_learning3.8.10scikit-learn=0.24.2

4. Database Information​

SQLite Databases​

  • dnv_coe_buildings.db: Buildings database
  • dnv_coe.db: LOE database (needs renaming)

5. Key Files and Directories​

Documentation​

  • readme.md: Project overview and setup instructions
  • training-and-predictions.md: Model training and prediction documentation
  • 02_loe-coe-update_7_29_2024.md: LOE COE update documentation

Implementation​

  • app.py: Streamlit application
  • script_runner.py: Original project implementation

Data​

  • data/ directory containing SQLite databases
  • Configuration files (.toml)

6. Dependencies and Requirements​

Python Versions​

  • 3.8.10
  • 3.9.16

Key Packages​

  • scikit-learn
  • streamlit

Version Constraints​

  • scikit-learn compatibility varies between environments
  • Python version dependencies must be carefully managed

7. Known Issues and Notes​

Technical Issues​

  1. Streamlit errors may lack detailed descriptions
  2. Python version compatibility issues with scikit-learn
  3. Database naming inconsistencies
  4. Functionality gaps in LOE calculations

Development Notes​

  • Regular updates to documentation are required
  • Environment setup requires careful version management
  • Database standardization needed

8. Maintenance and Updates​

Documentation Updates​

  • This document should be updated as new information becomes available
  • Version control should be maintained for all documentation changes

Environment Management​

  • Keep track of package version changes
  • Document any new environment setups
  • Maintain compatibility matrices

Note: This document serves as a living reference and should be updated as the project evolves.