MLOps NLP Ticket Classifier

A beginner-friendly MLOps project that classifies customer support tickets into business categories such as billing, delivery, refunds, technical support, and account issues. MLOps

#Machine Learning #DeepLearning #Python #scikit-learn #FastAPI #NLP #TextClassification #Data Science

Project Structure

• data/ - synthetic dataset storage
• src/ - source code
• models/ - trained models
• tests/ - unit tests
• docs/ - documentation

How to Run

1. Generate Dataset

python src/data/make_dataset.py

2. Train Model

python src/models/train_model.py

3. Run Tests

pythom -m pytest

Model Evaluation

After training, the model is evaluated on a test split (20% of data).

• Metrics: Accuracy, Precision, Recall, F1-score
• Confusion Matrix generated
• Full evaluation report saved in /reports/evaluation_report.txt

Example: Classification Report

               precision    recall  f1-score   support

      account       1.00      1.00      1.00        27
      billing       1.00      1.00      1.00        19
     delivery       1.00      1.00      1.00        22

Run API

Start the API server:

uvicorn src.api.app:app --reload

Health check:

Health Check

Swagger docs:

Doc