96 lines (66 loc) · 2.44 KB

Delivery ETA System Guide

This guide provides step-by-step instructions on how to operate the Delivery ETA system. Follow the steps below to process data, perform exploratory data analysis (EDA), train models, make predictions, and evaluate the results.

1. Data Processing

Script: `src/preprocessing.py`

Purpose: This script processes raw data and prepares it for analysis and modeling.
Input: Raw data files located in data/raw/.
Output: Processed data saved in data/processed/.
How to Run:
```
python src/preprocessing.py
```

2. Exploratory Data Analysis (EDA)

Script: `src/eda_summary.py`

Purpose: This script generates a summary of the dataset, including key statistics and visualizations.
Input: Processed data from data/processed/.
Output: EDA results saved in results/ (e.g., eda_summary.csv) and visualizations in results/figures/.
How to Run:
```
python src/eda_summary.py
```

3. Model Training

Script: `src/train_model.py`

Purpose: This script trains machine learning models using the processed data.
Input: Processed data from data/processed/.
Output: Trained model artifacts saved in models/.
How to Run:
```
python src/train_model.py
```

4. Making Predictions

Script: `src/predict.py`

Purpose: This script uses the trained model to make predictions on new data.
Input: New data or test data.
Output: Predictions saved in results/ (e.g., amazon_model_results.csv).
How to Run:
```
python src/predict.py
```

5. Model Evaluation

Script: `src/evaluate_model.py`

Purpose: This script evaluates the performance of the trained model using metrics such as accuracy, precision, recall, etc.
Input: Predictions and ground truth data.
Output: Evaluation metrics saved in results/tables/ (e.g., amazon_test_metrics.csv).
How to Run:
```
python src/evaluate_model.py
```

6. Deep Learning Model

Script: `src/dl_model.py`

Purpose: This script contains a deep learning model built from scratch for advanced predictions.
How to Run:
```
python src/dl_model.py
```

Notes

Ensure all dependencies are installed before running the scripts. Use the requirements.txt file to install them:
```
pip install -r requirements.txt
```