How kali-kar/student-performance-predictor works
Student Performance Predictor generates or loads student data, trains and evaluates regression models, saves the selected model artifacts, then accepts student details through a Flask form and returns a score, grade, and remark. The graph preserves the notebook-based EDA/training path documented in the README. The notebook, templates, styles, and serialized artifacts are represented from the tree and README; their source contents were not sampled.
Main entry points: app.py Flask application · train_model.py Train and evaluate · index.html Input and results · student_data.csv Student dataset
Components 12
Data
- Synthetic data
data/generate_data.py - Student dataset
data/student_data.csv
Model training
- EDA and training
notebook/Student_Performance_Prediction.ipynb - Train and evaluate
model/train_model.py
Saved artifacts
- Selected regressor
model/student_model.pkl - Feature scaler
model/scaler.pkl - Model metadata
model/model_meta.pkl
Web prediction
- Flask application
app/app.py - Input and results
app/templates/index.html - Web styling
app/static/style.css - Grade and remark
app/app.py
Other components
- Student
Connections 16
- Synthetic data → Student dataset: writes CSV
data/generate_data.py - Train and evaluate → Student dataset: reads data
model/train_model.py - EDA and training → Student dataset: loads or generates
README.md - EDA and training → Selected regressor: saves selected model
README.md - Train and evaluate → Selected regressor: saves selected model
model/train_model.py - Train and evaluate → Feature scaler: saves scaler
model/train_model.py - Train and evaluate → Model metadata: saves metadata
model/train_model.py - Student → Flask application: submits details
README.md - Flask application → Selected regressor: loads and predicts
app/app.py - Flask application → Feature scaler: loads and transforms
app/app.py - Flask application → Model metadata: loads features
app/app.py - Flask application → Input and results: renders form
app/app.py - Flask application → Grade and remark: calculates grade
app/app.py - Grade and remark → Input and results: renders result
app/app.py - Input and results → Web styling: uses styling
README.md - Flask application → Student: returns prediction
README.md