MoodLens-API

MoodLens-API is a FastAPI-based REST API for sentiment analysis of text.
It uses a trained scikit-learn model to classify text as positive or negative.
The API supports both single and batch text predictions and is designed to integrate easily into chatbots, web apps, or other NLP pipelines.

---

Features

• Predict sentiment for a single text input
• Predict sentiment for a batch of texts
• Returns predicted sentiment and confidence score
• Built with FastAPI for high performance and scalability
• Easy to deploy on platforms like Render, Heroku, or AWS

---

Project Structure

moodlens-api/ │── sent_app.py # FastAPI app code
│── requirements.txt # Python dependencies
│── .gitignore # Ignore ML models, env files, etc.
│── README.md # Project description
│── start.sh # Optional: script to start the API

Note: ML model files (.joblib) are not included in the repo to keep it lightweight.
Download models separately or store them in cloud storage for production.

---

Installation

1. Clone the repository: bash
git clone https://github.com/nyarderr/moodlens-api.git
cd moodlens-api
   

2. Install dependencies pip install -r requirements.txt
   

3. Place your trained model files (model.joblib, vectorizer.joblib, label_encoder.joblib) in the project root.
   

---

Running the API

Locally

bash
uvicorn sent_app:app --reload
Visit http://127.0.0.1:8000/docs for interactive API docs (Swagger UI).

API EndPoints

1. Health Check
   bash
GET /health
   Response;
   json
{"status":"ok,"model":"loaded"}
   

2. Predict Sentiment
   bash
POST /predict
   PayLoad;
   json
{"text":"I love this product"}- single text OR
   {"text": ["I love this!", "This is terrible."]} - batch input (list of texts)
   Response;
   json
{"Sentiment":"positive","Confidence:0.92} OR
   [{"text": "I love this!", "Sentiment":"positive", "Confidence":0.93},{"text": "This is terrible.", "Sentiment":"negative", "Confidence":0.83}]
   

---

Deployment

• Recommended platforms: Render, Heroku, AWS, GCP

• Use uvicorn sent_app:app --host 0.0.0.0 --port $PORT as the start command.

• Ensure model files are accessible in production (local or cloud storage).