praven-pro

Praven Pro - Interface Screenshots

Visual guide to the Praven Pro user interfaces.

Web Interface (port 5001)

Landing Page

Web Interface - Landing Page

Features shown:

URL: http://localhost:5001

File Upload

Web Interface - File Upload

Features shown:

Validation Results

Web Interface - Results

Features shown:

Validation Dashboard

Overview Section

Dashboard - Overview

Features shown:

Validation Breakdown Chart

Dashboard - Validation Chart

Features shown:

Species Summary Table

Dashboard - Species Table

Features shown:

Rejection Reasons

Dashboard - Rejection Reasons

Features shown:

Smart Review Section

Dashboard - Smart Review

Features shown:

Command Line Interface

Basic Validation Command

$ python validate.py gaulossen_all_detections.csv \
    --lat 63.341 --lon 10.215 \
    --habitat wetland \
    --date 2025-10-15

Loading BirdNET results: gaulossen_all_detections.csv
  Loaded 6,805 detections
  Unique species: 85

Checking eBird cache...
✓ Cache found: cache/ebird_63.341_10.215_50km.json
✓ Cache age: 2 days (fresh, using cached data)

Running biological validation...
[████████████████████████████████████] 6,805/6,805 (100%)

Smart Review Selection
================================================================================
Total detections: 6,805
  Auto-accepted: 581 (no review needed)
  Auto-rejected: 23 (no review needed)
  Needs review: 6,201

📋 Review Workload Reduction:
  Before: 6,201 detections to review
  After:  192 detections to review
  Reduction: 96.9%

  Species with REVIEW status: 64
  Top 3 per species selected

Exported full results: output/praven_results_20251022_032840_full.csv
Exported accepted: output/praven_results_20251022_032840_accepted.csv
Exported rejected: output/praven_results_20251022_032840_rejected.csv
Exported review: output/praven_results_20251022_032840_review.csv

🎯 Smart Review Selection:
   Priority review (top 3/species): output/praven_results_20251022_032840_PRIORITY_REVIEW.csv
   → Review only 192 detections instead of 6,201!
   → 97% workload reduction

================================================================================
Validation Complete!
================================================================================

Results:
  Total:        6,805 detections
  Accepted:     581 (8.5%)
  Rejected:     23 (0.3%)
  Needs Review: 6,201 (91.1%)

Species:
  Accepted: 19 species
  Rejected: 5 species

🎯 Smart Review Reduction:
  Original workload: 6,201 detections
  Priority review:   192 detections (top 3/species)
  Workload saved:    97%
  → Review the PRIORITY_REVIEW.csv file instead of review.csv!

Help Menu

$ python validate.py --help

usage: validate.py [-h] --lat LAT --lon LON --habitat
                   {wetland,forest,oceanic,grassland,urban,agricultural}
                   [--date DATE] [--rain RAIN] [--fog FOG] [--no-ebird]
                   input_csv

Praven Pro - Automated BirdNET Validation

positional arguments:
  input_csv             Path to BirdNET detections CSV

optional arguments:
  -h, --help            show this help message and exit
  --lat LAT             Latitude of study location
  --lon LON             Longitude of study location
  --habitat {wetland,forest,oceanic,grassland,urban,agricultural}
                        Primary habitat type
  --date DATE           Study date (YYYY-MM-DD format)
  --rain RAIN           Rain intensity (0.0-1.0)
  --fog FOG             Fog density (0.0-1.0)
  --no-ebird            Disable eBird integration

Examples:
  # Basic validation
  python validate.py detections.csv --lat 63.341 --lon 10.215 --habitat wetland

  # With weather conditions
  python validate.py detections.csv --lat 63.341 --lon 10.215 --habitat wetland \
    --rain 0.3 --fog 0.5 --date 2025-10-15

  # Without eBird (faster, less accurate)
  python validate.py detections.csv --lat 63.341 --lon 10.215 --habitat wetland \
    --no-ebird

Output Files

Priority Review CSV

Priority Review CSV

Columns shown:

Sample rows:

common_name,confidence,quality_score,temporal_valid,habitat_valid,status
Graylag Goose,0.95,1.05,true,true,REVIEW
Graylag Goose,0.92,1.02,true,true,REVIEW
Graylag Goose,0.88,0.98,true,true,REVIEW
Great Snipe,0.87,0.97,true,false,REVIEW
Great Snipe,0.83,0.93,true,false,REVIEW
Great Snipe,0.79,0.89,true,false,REVIEW

Validation Dashboard HTML

Dashboard HTML

Features:

File location: output/praven_results_YYYYMMDD_HHMMSS_dashboard.html

Screenshots Capture Guide

To capture these screenshots:

Web Interface

  1. Start web server: python web_app.py
  2. Open browser: http://localhost:5001
  3. Capture landing page
  4. Upload test CSV file
  5. Fill form with test data
  6. Capture results page
  7. Save screenshots to docs/images/web_*.png

Dashboard

  1. Open generated dashboard HTML in browser
  2. Capture full page overview
  3. Capture individual sections (charts, tables)
  4. Save to docs/images/dashboard_*.png

Command Line

  1. Use terminal screenshot tool (macOS: Cmd+Shift+4)
  2. Run validation commands
  3. Capture output
  4. Save to docs/images/cli_*.png

CSV Files

  1. Open in Excel or Google Sheets
  2. Format for readability
  3. Capture header and sample rows
  4. Save to docs/images/csv_*.png

File Naming Convention

docs/images/
├── web_interface_landing.png       # Web UI landing page
├── web_interface_upload.png        # File upload area
├── web_interface_results.png       # Results page
├── dashboard_overview.png          # Dashboard main view
├── dashboard_validation_chart.png  # Validation breakdown chart
├── dashboard_species_table.png     # Species summary table
├── dashboard_rejection_reasons.png # Rejection reasons chart
├── dashboard_smart_review.png      # Smart review section
├── priority_review_csv.png         # Priority review CSV sample
├── dashboard_html.png              # Dashboard HTML file
├── cli_validation.png              # CLI validation output
└── cli_help.png                    # CLI help menu

Screenshot Specifications

Format: PNG (lossless) Resolution: 1920x1080 minimum (or native retina) Compression: Optimized for web (use tools like ImageOptim) Annotations: Optional - highlight key features with arrows/boxes Naming: Lowercase, underscores, descriptive

Future Screenshots

As new features are added:

  1. Capture screenshots following this guide
  2. Add to docs/images/ with descriptive names
  3. Update this document with new sections
  4. Link screenshots in main README.md

Notes

This site is open source. Improve this page.