How lavjaiswal034/student_performence_prediction works
This is a student-performance prediction app: students submit academic and engagement data through the web frontend, the Flask API predicts a category, adds risk reasons and a tailored study path, then records the result for history. A separate model-development workflow generates training data and compares Random Forest with XGBoost; the saved model and encoders are loaded by the API. The repository also includes EDA, model-comparison charts, and SHAP summaries. Runtime wiring is shown in the excerpts; the training and reporting relationships rely on their own excerpts and README workflow.
Main entry points: app.py Prediction API · student_data.csv Student Dataset · script.js Input & Results UI · student_model.pkl Model & Encoders
Components 14
Student App
- Input & Results UI
frontend/script.js - Prediction API
backend/app.py
Model Development
- Student Dataset
dataset/student_data.csv - Synthetic Data Generator
ml/generate_synthetic_dataset.py - Model Training & Comparison
ml/train_model.py - Model & Encoders
ml/student_model.pkl - Performance Prediction
backend/app.py
Explanation & Guidance
- Risk Reason Generation
src/explainability.py - Personalized Study Path
src/learning_path.py
History & Analysis
- Prediction History Store
src/database.py - Exploratory Analysis
ml/generate_eda.py - Model & Factor Charts
ml/generate_simple_graphs.py - SHAP Summary
ml/generate_shap_summary.py
Other components
- Student
Connections 16
- Student → Input & Results UI: enters data
frontend/script.js - Input & Results UI → Prediction API: posts input
frontend/script.js - Prediction API → Model & Encoders: loads model
backend/app.py - Prediction API → Performance Prediction: predicts
backend/app.py - Prediction API → Risk Reason Generation: generates reasons
backend/app.py - Prediction API → Personalized Study Path: builds study path
backend/app.py - Prediction API → Prediction History Store: saves result
backend/app.py - Prediction API → Prediction History Store: gets history
backend/app.py - Prediction API → Input & Results UI: returns result
frontend/script.js - Synthetic Data Generator → Student Dataset: writes data
ml/generate_synthetic_dataset.py - Model Training & Comparison → Student Dataset: reads data
ml/train_model.py - Model Training & Comparison → Model & Encoders: saves model
ml/train_model.py - Exploratory Analysis → Student Dataset: analyzes data
ml/generate_eda.py - Model & Factor Charts → Student Dataset: charts categories
ml/generate_simple_graphs.py - SHAP Summary → Student Dataset: explains features
ml/generate_shap_summary.py - SHAP Summary → Model & Encoders: loads model
ml/generate_shap_summary.py