Skip to content

README.md

Latest commit

 

History

History
106 lines (85 loc) · 4.25 KB

README.md

File metadata and controls

106 lines (85 loc) · 4.25 KB

Scripts Directory

This directory contains Python scripts for data processing, analysis, and figure generation.

Directory Structure

scripts/
├── analysis/         # Data collection and processing scripts
│   ├── collect_dependency_data.py     # Fetch PyPI/conda-forge metadata
│   ├── process_dependency_data.py     # Clean and aggregate dependency data
│   └── ...
└── plotting/         # Figure generation scripts (12 scripts)
    ├── generate_architecture_diagram.py      # 12 Terraform-based diagrams
    ├── generate_sleap_dependency_graph.py    # SLEAP dependency network
    ├── generate_qr_codes.py                   # QR codes for demos
    ├── plot_software_complexity.py            # Dependency growth over time
    ├── plot_gpu_cost_trends.py                # GPU pricing trends
    ├── plot_gpu_reliance.py                   # GPU dependency scoring
    ├── plot_deployment_impact.py              # Workshop timeline
    ├── plot_os_distribution.py                # OS analysis
    ├── plot_configuration_hierarchy.py        # LabLink config visualization
    ├── plot_configuration_hierarchy_simple.py # Simplified config diagram
    ├── gpu_cost_analysis.py                   # LabLink session cost breakdown
    └── plot_lablink_maintainability.py        # LabLink code maintainability metrics

Script Categories

Analysis Scripts (scripts/analysis/)

Purpose: Collect and process raw data for analysis figures.

Data collection and processing with caching, API rate limiting, and validation.

Example workflow:

cd scripts/analysis
uv run python collect_dependency_data.py --github-token YOUR_TOKEN
uv run python process_dependency_data.py
# Output: data/processed/dependency_data.csv

Plotting Scripts (scripts/plotting/)

Purpose: Generate publication-quality figures with consistent styling.

All scripts support:

  • --preset {paper|poster|presentation} - Font size presets
  • --format {png|pdf|svg|both|all} - Output formats
  • --output-dir - Output location
  • --verbose - Debug logging

Example:

cd scripts/plotting
uv run python plot_gpu_cost_trends.py --preset poster --format pdf

Common Usage Patterns

Font Size Presets

--preset paper          # 14pt fonts (default)
--preset poster         # 20pt fonts (large format)
--preset presentation   # 16pt fonts (slides)

Output Formats

--format png      # 300 DPI raster (default)
--format pdf      # Vector for LaTeX
--format both     # PNG + PDF
--format all      # PNG + PDF + SVG

Environment Variables

export LABLINK_TERRAFORM_DIR=../lablink-template/lablink-infrastructure
export SLEAP_PATH=../sleap
export GITHUB_TOKEN=ghp_your_token

All Plotting Scripts

  1. generate_architecture_diagram.py - 12 Terraform-based infrastructure diagrams
  2. generate_sleap_dependency_graph.py - SLEAP dependency network
  3. generate_qr_codes.py - QR codes for demos
  4. plot_software_complexity.py - Dependency growth (2000-2025)
  5. plot_gpu_cost_trends.py - GPU pricing trends (2006-2025)
  6. plot_deployment_impact.py - Workshop timeline
  7. plot_os_distribution.py - OS analysis
  8. plot_gpu_reliance.py - GPU dependency scoring
  9. plot_configuration_hierarchy.py - Config visualization (detailed)
  10. plot_configuration_hierarchy_simple.py - Config visualization (simple)
  11. gpu_cost_analysis.py - LabLink infrastructure cost breakdown for a tutorial session (table, vertical stacked bar, horizontal stacked bar). Supports --preset {paper,poster,presentation}, --format {png,pdf,both}, --vms, --hours, and writes a metadata sidecar.
  12. plot_lablink_maintainability.py - Per-package cyclomatic complexity, LOC, and maintainability index (via radon) for LabLink. Writes PNG + PDF plus a metadata sidecar.

See Figure Documentation for detailed usage of each script.

Related Documentation