How muhil-amuthan/food-delivery-time-prediction works
This repository is a food-delivery-time prediction project. Its main workflow is to load and clean the delivery dataset, encode features, compare regression models, and save the selected model and encoders; a Streamlit interface then accepts delivery details and displays a prediction. The training script also performs exploratory analysis. The dataset and serialized artifacts are shown as files; their contents were not sampled.
Main entry points: Food_Delivery_Prediction.py Clean and encode · app.py Input interface · delivery_model.pkl Saved model · encoders.pkl Saved encoders
Components 11
Training and analysis
- Delivery dataset
Food_Delivery_Times.csv - Clean and encode
Food_Delivery_Prediction.py - Explore data
Food_Delivery_Prediction.py - Train regressors
Food_Delivery_Prediction.py - Compare models
Food_Delivery_Prediction.py - Saved model
delivery_model.pkl - Saved encoders
encoders.pkl
Prediction app
- Input interface
app.py - Encode and predict
app.py - Estimated time
app.py
Other components
- User
Connections 11
- Clean and encode → Delivery dataset: reads
Food_Delivery_Prediction.py - Clean and encode → Explore data: prepares data for
Food_Delivery_Prediction.py - Clean and encode → Train regressors: provides features to
Food_Delivery_Prediction.py - Train regressors → Compare models: evaluated by
Food_Delivery_Prediction.py - Compare models → Saved model: saves selected model
Food_Delivery_Prediction.py - Clean and encode → Saved encoders: saves encoders
Food_Delivery_Prediction.py - User → Input interface: enters details
app.py - Input interface → Saved model: loads
app.py - Input interface → Saved encoders: loads
app.py - Input interface → Encode and predict: submits inputs
app.py - Encode and predict → Estimated time: displays prediction
app.py