🧠 Sentiment & Image Classification Microservices

This project consists of multiple Dockerized microservices to perform sentiment prediction and image classification via deep learning models, accessible through a single frontend UI.

---

📥 Clone the Repository

git clone https://github.com/LCIT-AISC-T3-S25/Group5.git
cd Group5

---

🚀 Run the Entire Project

Use Docker Compose to build and run all services with a single command:

docker-compose up --build

This will spin up the following services:

Service	Port	Description
UI	5002	HTML frontend (Flask)
BiLSTM GRU	5004	Sentiment prediction (GRU)
Word2Vec LSTM	5003	Sentiment prediction (Word2Vec)
VGG16 FastAPI	5005	Image classification using VGG16
CNN FastAPI	5006	Custom CNN model for image input

---

🛠️ Manual Build & Run (if not using Compose)

BiLSTM GRU (Port 5004)

docker build -t bi_lstm ./GRU_Sentiment
docker run -v $(pwd):/app -p 5004:5004 bi_lstm

Word2Vec LSTM (Port 5003)

docker build -t word2vec ./word2vec_lstm_Model
docker run -v $(pwd):/app -p 5003:5003 word2vec

UI (Port 5002)

docker build -t models_html ./HTML/flask-app-container
docker run -v $(pwd):/app -p 5002:5002 models_html

VGG16 FastAPI (Port 5005)

docker build -t vgg16_fast_api ./VGG16_FastAPI
docker run -v $(pwd):/app -p 5005:5005 vgg16_fast_api

CNN FastAPI (Port 5006)

docker build -t cnn ./CNN_FastAPI
docker run -v $(pwd):/app -p 5006:5006 cnn

---

🌐 Access the Application

Once all containers are running, open your browser and navigate to:

http://localhost:5002

---

📂 Project Structure

Group5/
│
├── docker-compose.yml
├── HTML/flask-app-container # Flask UI (Port 5002)
├── GRU_Sentiment/           # BiLSTM GRU model (Port 5004)
├── word2vec_lstm_Model/     # Word2Vec LSTM model (Port 5003)
├── VGG16_FastAPI/           # VGG16 model in FastAPI (Port 5005)
└── CNN_FastAPI/             # CNN model in FastAPI (Port 5006)

---

🧪 Health Check Endpoints

You can check if services are running with:

• http://localhost:5002/health
• http://localhost:5003/health
• http://localhost:5004/health
• http://localhost:5005/docs (VGG16 FastAPI Swagger UI)
• http://localhost:5006/docs (CNN FastAPI Swagger UI)

---

📦 Requirements

• Docker
• Docker Compose

---

📜 License

MIT © 2025 Group5