How karunya2006-ux/california_house_price_predictoin works
This California housing predictor offers two related workflows: the Streamlit app reads census data, trains a model at startup, and predicts from submitted property details; the separate script trains and serializes model artifacts, or loads them to predict on a test split. The README also documents stratified sampling, preprocessing, Random Forest training, and hosted Streamlit use. The graph reflects the supplied source and tree; serialized files and deployment hosting are documented or runtime outputs, not present in the tree.
Main entry points: main.py Training script · app.py Streamlit app · housing.csv Census dataset
Components 13
Prediction app
- Streamlit app
app.py - Stratified split
app.py - Feature preprocessing
app.py - Random Forest model
app.py - Price estimate
app.py
Training workflow
- Training script
main.py - Stratified split
main.py - Pipeline builder
main.py - Random Forest model
main.py - Serialized artifacts
- Prediction CSV
Housing data
- Census dataset
housing.csv
Other components
- Property user
Connections 13
- Property user → Streamlit app: enters features
- Streamlit app → Census dataset: reads
- Streamlit app → Stratified split: samples training data
- Streamlit app → Feature preprocessing: fits and transforms
- Streamlit app → Random Forest model: fits and predicts
- Streamlit app → Price estimate: displays
- Training script → Census dataset: reads
- Training script → Stratified split: stratifies
- Training script → Pipeline builder: builds pipeline
- Training script → Random Forest model: fits model
- Training script → Serialized artifacts: serializes
- Training script → Serialized artifacts: loads
- Training script → Prediction CSV: writes predictions