📊 Predict Sales AI Model

A simple machine learning project to predict future sales totals based on product, customer, quantity, and price using a Random Forest Regressor and Linear Regression. This project includes:

• Data preprocessing & model training
• Flask API for real-time predictions
• Visualization of prediction performance
• Easy setup & deployment-ready code

🔮 Test API

https://3c96b8b2-4475-48a0-941c-69ab9e588cff-00-1m3398d51iq58.pike.replit.dev:5000

OR

https://sales-predictor-ai-engine-production.up.railway.app/api/v1/predict

✅ Batch Input Example

curl -X POST https://3c96b8b2-4475-48a0-941c-69ab9e588cff-00-1m3398d51iq58.pike.replit.dev:5000/api/v1/predict  -H "Content-Type: application/json"  -d '[
       {"product": "Laptop", "customer": "John", "quantity": 2, "price": 800},
       {"product": "Tablet", "customer": "Alice", "quantity": 3, "price": 400},
       {"product": "Phone", "customer": "Mark", "quantity": 0, "price": 900}
     ]'

Response:

[
  {
    "index": 0,
    "predicted_total": 1520.0
  },
  {
    "index": 1,
    "predicted_total": 1205.0
  },
  {
    "index": 2,
    "error": "Quantity must be a positive number."
  }
]

✅ Valid records will also be appended to data/new_sales.csv for future retraining purposes.

---

🗂️ Project Structure

predict_sales_ai_model/
├── api/
│    ├── routes
|    |    └──__init__.py
|    |    └──v1.py
|    ├── _init_.py
|    └── app.py
├── charts/
│   └── random_forest_prediction_vs_actual.png
├── data/
│   └── sales_50.csv
│   └── sales_1000.csv
├── model/
│   └──random_forest
|       ├── model.pkl
|       ├── model_columns.pkl
|       └── train_and_save.py
├── tests/
│   └── test_random_forest_model.py (optional)
├── requirements.txt
├── .gitignore
└── README.md

---

🔍 Features Used

• product: Product name (categorical)
• customer: Customer name (categorical)
• quantity: Number of units sold
• price: Price per unit

total = quantity × price is used as the prediction target.

---

🧠 Model Details

• Algorithm: Random Forest Regressor

• Framework: scikit-learn

• Evaluation:

  • Mean Absolute Error: ~19.82 AED
  • R² Score: 1.0 (excellent fit)

---

🧪 Training the Model

# Activate your virtual environment
python -m venv venv
venv\Scripts\activate  # On Windows

# Install dependencies
python -m pip install -r requirements.txt

# Train and save the model
python -m model.random_forest.train_and_save

---

🧪 Testing the Model

# Test the model
python -m tests.test_random_forest_model

---

🔮 Running the Flask API Local

python -m api.app

The API will start at: http://127.0.0.1:5000/

✅ Single Input Example

curl -X POST http://127.0.0.1:5000/api/v1/predict -H "Content-Type: application/json"  -d '{"product": "Laptop", "customer": "John", "quantity": 2, "price": 800}'

Response:

[
  {
    "index": 0,
    "predicted_total": 1520.0
  }
]

✅ Batch Input Example

curl -X POST http://127.0.0.1:5000/api/v1/predict  -H "Content-Type: application/json"  -d '[
       {"product": "Laptop", "customer": "John", "quantity": 2, "price": 800},
       {"product": "Tablet", "customer": "Alice", "quantity": 3, "price": 400},
       {"product": "Phone", "customer": "Mark", "quantity": 0, "price": 900}
     ]'

Response:

[
  {
    "index": 0,
    "predicted_total": 1520.0
  },
  {
    "index": 1,
    "predicted_total": 1205.0
  },
  {
    "index": 2,
    "error": "Quantity must be a positive number."
  }
]

✅ Valid records will also be appended to data/new_sales.csv for future retraining purposes.

---

📈 Prediction Visualization

This chart compares actual and predicted sales totals on the test set.

---

📋 Requirements

Flask
pandas
scikit-learn
joblib
matplotlib

---

🚫 .gitignore

__pycache__/
*.pkl
.env
venv/
*.pyc
.DS_Store

---

📝 License

MIT License

---

🙌 Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

---

👨‍💻 Author

Shahzad Software Engineer | ML Enthusiast

---

For any questions or suggestions, feel free to reach out!