Skip to content

demo_phase2.md

Latest commit

 

History

History
270 lines (198 loc) · 6.39 KB

demo_phase2.md

File metadata and controls

270 lines (198 loc) · 6.39 KB

Phase 2 Demo Script

Demo Duration: 7-10 minutes

This script guides you through demonstrating the Phase 2 features of IntelliWeather API.

Setup (Before Demo)

# Start the server
cd Weather-API
pip install -r requirements.txt
uvicorn app:app --reload --port 8000

# Open Swagger docs
open http://localhost:8000/docs

Part 1: Weather V2 Endpoints (2 min)

Show Enhanced Weather API

  1. Open Swagger UI at http://localhost:8000/docs

    • Point out the organized tags: Weather V2, Geocoding, Alerts, etc.

  2. Hourly Forecast

    curl "http://localhost:8000/weather/hourly?lat=40.7128&lon=-74.006&hours=48"
    • Show the structured JSON response
    • Highlight: temperature, humidity, precipitation probability

  3. CSV Export

    curl "http://localhost:8000/weather/hourly?lat=40.7128&lon=-74.006&format=csv" -o hourly.csv
open hourly.csv
    • "Users can export data to CSV for their own analysis"

  4. Historical Data

    curl "http://localhost:8000/weather/historical?lat=40.7128&lon=-74.006&start=2024-01-01&end=2024-01-07"
    • "We can fetch historical weather for any date range up to 1 year"

Part 2: Geocoding & Search (1 min)

Location Search

  1. Search for a city

    curl "http://localhost:8000/geocode/search?q=Lahore&limit=3"
    • Show coordinates, country, timezone, population

  2. Reverse Geocoding

    curl "http://localhost:8000/geocode/reverse?lat=31.5204&lon=74.3587"
    • "From coordinates, we can get the location name"

  3. Show Cache Hit

    • Make the same request again
    • Point out "source": "cache" in response
    • "Second request is instant - served from cache"

Part 3: Weather Alerts (1 min)

Alert System

  1. Get Alerts

    curl "http://localhost:8000/alerts?lat=40.7128&lon=-74.006"
    • Show alert structure: severity, headline, instructions

  2. Explain the System

    • "Alerts are generated based on weather conditions"
    • "In production, this would integrate with National Weather Service"
    • "Alerts are cached for 15 minutes"

Part 4: Downloadable Reports (1.5 min)

PDF and Excel Reports

  1. Download PDF Report

    curl "http://localhost:8000/weather/download?lat=40.7128&lon=-74.006&type=pdf&range=daily&days=7" -o report.pdf
open report.pdf
    • Show the formatted report with temperature chart

  2. Download Excel Report

    curl "http://localhost:8000/weather/download?lat=40.7128&lon=-74.006&type=excel&range=daily&days=7" -o report.xlsx
open report.xlsx
    • Show multiple sheets: Weather data, Metadata
    • "Excel files have proper styling and formatting"

Part 5: ML Prediction (1 min)

Temperature Prediction

  1. Get Next-Day Prediction

    curl "http://localhost:8000/predict/nextday?lat=40.7128&lon=-74.006"
    • Show predicted temperature and confidence interval
    • "Simple ML model using historical data"

  2. Model Info

    curl "http://localhost:8000/predict/model/info"
    • Show model version and training date

Part 6: Multi-Language Support (1 min)

Internationalization

  1. Show Supported Languages

    curl "http://localhost:8000/i18n/languages"
    • List: English, Hindi, Urdu, Arabic, Spanish

  2. Get Hindi Translations

    curl "http://localhost:8000/i18n/translations?lang=hi" | jq '.translations | {current_weather, feels_like, humidity}'
    • Show Hindi text

  3. Weather Description in Arabic

    curl "http://localhost:8000/i18n/weather-description?code=3&lang=ar"
    • Show Arabic translation

Part 7: API Key System (1 min)

API Key Management

  1. Login First

    curl -X POST "http://localhost:8000/auth/login" \
  -H "Content-Type: application/json" \
  -d '{"email":"demo@example.com","password":"demo123"}' \
  -c cookies.txt

  2. Create API Key

    curl -X POST "http://localhost:8000/apikeys" \
  -H "Content-Type: application/json" \
  -d '{"name":"Demo Key","rate_limit":100}' \
  -b cookies.txt
    • "Key is only shown once - store securely"

  3. Use API Key

    curl "http://localhost:8000/weather/hourly?lat=40.7128&lon=-74.006" \
  -H "X-API-Key: YOUR_API_KEY"
    • "Each key has its own rate limit"

Part 8: Admin Dashboard (1 min)

Analytics Dashboard

  1. Open Dashboard

    • Navigate to http://localhost:8000/admin/dashboard
    • (Requires admin user)

  2. Show Statistics

    • Total users
    • Active sessions
    • Cache hit rate
    • Top searched locations

  3. System Health

    • Storage status
    • Cache operational
    • API connectivity

Part 9: Scaling Demo (Optional, 1 min)

Docker Scaling

# Show docker-compose
cat docker-compose.yml

# Scale to 3 instances
docker-compose up -d --scale app=3

# Show NGINX load balancing
curl http://localhost/weather?lat=40.7128&lon=-74.006

Closing Points

  1. Architecture Highlights

    • Multi-tier caching (memory → CSV → external API)
    • Feature flags for all Phase 2 features
    • Structured JSON logging
    • Thread-safe operations

  2. Production Ready

    • Docker support with load balancing
    • Health checks and metrics
    • Rate limiting per IP and per API key
    • Secure session management

  3. Documentation

    • Swagger/OpenAPI docs
    • README.md
    • QA checklist
    • Demo scripts

Troubleshooting

Issue Solution
Server won't start Check pip install -r requirements.txt
401 errors Login first or provide API key
429 errors Rate limit hit, wait 60 seconds
Empty alerts No severe weather at that location
PDF not rendering Check if data fetched successfully

Q&A Preparation

Q: Why CSV storage instead of a database? A: Simpler setup, no external dependencies. Demonstrates file-based persistence. Production would use PostgreSQL/Redis.

Q: How does the ML model work? A: Linear regression on historical temperatures. Features: day of year (cyclical), previous temps, trend.

Q: How accurate are predictions? A: Simple model with ±3°C confidence. Production would use more sophisticated approaches.

Q: What happens if Open-Meteo is down? A: Cached data is returned if available. Error responses include 503 status with clear message.