🗣️ Sentiment Analyzer API

📌 Overview

A Machine Learning API for real-time sentiment analysis using scikit-learn and FastAPI. This project demonstrates core NLP techniques, robust engineering practices, and automated CI/CD with GitHub Actions.

---

🎯 Objectives

• Train a text classification model to predict sentiment (positive, negative, neutral).
• Automate linting, testing, and deployment with GitHub Actions.

---

📊 Dataset

Uses the IMDB Movie Reviews Dataset or any dataset of labeled text.

Example features:

• review_text
• label (positive, negative)

---

🧠 ML Concepts Implemented

1️⃣ Text Vectorization

• Converts raw text to numeric features using CountVectorizer or TfidfVectorizer.

2️⃣ Model Training

• Uses Logistic Regression or Support Vector Machine (SVM) for classification.

3️⃣ Evaluation

• Train/test split, accuracy, and classification report.
• Save trained model and vectorizer with joblib.

---

⚙️ DevOps Integration

🚀 GitHub Actions CI/CD

Automates:

• Linting (flake8)
• Unit tests for model & API
• Coverage reporting

---

📚 Deliverables

• ✅ Clean Python scripts for training and serving.
• ✅ Jupyter Notebook for exploration and training.
• ✅ CI/CD workflow with linting & tests.
• ✅ README with full instructions.

---

🚀 Usage

1️⃣ Install dependencies

pip install -r requirements.txt