How preethisomasundaram/menstracare works
MenstraCare is a PCOS prediction and model-evaluation web app: users enter clinical features, choose a trained model, and receive a prediction and probability; a separate results page displays evaluation metrics and visualizations. The graph follows the documented browser/API workflow and the sampled FastAPI and JavaScript. Model artifacts and results metadata are documented as backend inputs; their individual contents and page markup were not sampled.
Main entry points: app.py FastAPI app and static host · prediction.js Prediction form · training-results.js Results dashboard · models Feature and target metadata
Components 14
Web interface
- Prediction form
frontend/js/prediction.js - Results dashboard
frontend/js/training-results.js - Theme and navigation
frontend/js/common.js
Prediction service
- FastAPI app and static host
backend/app.py - Form schema and model list
backend/app.py - Prediction endpoint
backend/app.py
Model evaluation
- Training-results endpoint
backend/app.py - Metrics and chart rendering
frontend/js/training-results.js
Model artifacts
- Trained pipelines
backend/models - Deployment configuration
backend/models/deployment_config.json - Feature and target metadata
backend/models - Model performance metrics
backend/models/model_performance.json - Evaluation visual data
backend/models/training_artifacts.json
Other components
- User
Connections 19
- User → Prediction form: enters features
frontend/js/prediction.js - Prediction form → Form schema and model list: requests schema
frontend/js/prediction.js - Form schema and model list → Feature and target metadata: reads metadata
backend/app.py - Form schema and model list → Deployment configuration: reads model list
backend/app.py - Prediction form → Prediction endpoint: posts model and features
frontend/js/prediction.js - Prediction endpoint → Trained pipelines: loads and invokes pipeline
backend/app.py - Prediction endpoint → Feature and target metadata: validates and orders inputs
backend/app.py - Prediction endpoint → Prediction form: returns prediction
frontend/js/prediction.js - User → Results dashboard: opens results
README.md - Results dashboard → Training-results endpoint: requests evaluation data
frontend/js/training-results.js - Training-results endpoint → Model performance metrics: reads metrics
backend/app.py - Training-results endpoint → Evaluation visual data: reads visual data
backend/app.py - Training-results endpoint → Metrics and chart rendering: returns evaluation data
frontend/js/training-results.js - Metrics and chart rendering → Results dashboard: renders dashboard
frontend/js/training-results.js - FastAPI app and static host → Deployment configuration: loads configuration
backend/app.py - FastAPI app and static host → Trained pipelines: loads models
backend/app.py - FastAPI app and static host → Feature and target metadata: loads metadata
backend/app.py - FastAPI app and static host → Model performance metrics: loads metrics
backend/app.py - FastAPI app and static host → Evaluation visual data: loads artifacts
backend/app.py