How manusaviour1506/salary_prediction works
This is a Streamlit salary-prediction app: users enter experience, education, company size, country, and skills, then receive a model prediction. The runtime path assembles one-hot feature columns and calls the loaded model. The README also describes data collection, preprocessing, feature engineering, training, and evaluation, but their implementation is not shown; the notebook is present but unsampled, so those stages are documented rather than verified. Edges cite the app or README where the relationship is visible.
Main entry points: app.py Feature assembly · salarypred.ipynb Preprocessing · salary_model.pkl Salary model · columns.pkl Feature columns
Components 12
User interaction
- Input form
app.py - Predict action
app.py - Salary result
app.py
Prediction runtime
- Feature assembly
app.py - Feature columns
columns.pkl - Salary model
salary_model.pkl - Model inference
app.py
Model development
- Data collection
salarypred.ipynb - Preprocessing
salarypred.ipynb - Feature engineering
salarypred.ipynb - Training and evaluation
salarypred.ipynb
Other components
- User
Connections 11
- User → Input form: enters details
app.py - Input form → Predict action: submits
app.py - Predict action → Feature assembly: assembles features
app.py - Feature assembly → Feature columns: uses columns
app.py - Feature assembly → Model inference: passes features
app.py - Model inference → Salary model: predicts with
app.py - Model inference → Salary result: displays prediction
app.py - Data collection → Preprocessing: precedes
README.md - Preprocessing → Feature engineering: precedes
README.md - Feature engineering → Training and evaluation: precedes
README.md - Training and evaluation → Salary model: saves model
README.md