Skip to content

A collection of labs in deep learning and Flutter: CNN-based pneumonia detection, LSTM stock forecasting, image classification, and a smart mobile app with AI features using TensorFlow, FastAPI, and local LLMs.

Notifications You must be signed in to change notification settings

Youss2f/Deep-Learning-Portfolio

Repository files navigation

Deep Learning Portfolio

A comprehensive portfolio showcasing Deep Learning and Flutter development skills. This project demonstrates end-to-end ML workflows from model training to mobile deployment with an Agentic AI backend.

Deep Learning Flutter Python Firebase


Project Structure

├── Lab1_Image_Classification/   # CNN Image Classifier (Fashion MNIST)
├── Lab2_Stock_Prediction/       # LSTM Stock Price Predictor (Streamlit)
├── Lab3_Model_Deployment/       # Pneumonia X-Ray Classifier (Streamlit)
├── Lab4_Smart_App_Flutter/      # Flutter Mobile App with AI Features
├── Control_Backend_AgenticAI/   # MCP Server for Agentic AI
└── docker-compose.yaml          # Ollama + Open WebUI

Labs Overview

Lab Technology Description
Lab 1 TensorFlow/Keras (CNN) Image classification with Fashion MNIST
Lab 2 TensorFlow/Keras (LSTM) Stock price prediction with RNN
Lab 3 Streamlit + TensorFlow Pneumonia detection from X-rays
Lab 4 Flutter + Firebase Smart mobile app with AI chatbot

Quick Start

Prerequisites

  • Python 3.10+
  • Flutter 3.x
  • Docker (for Ollama)

Run Labs

# Lab 1: Image Classification
cd Lab1_Image_Classification
pip install -r requirements.txt
python train_model.py

# Lab 2: Stock Prediction
cd Lab2_Stock_Prediction
pip install -r requirements.txt
streamlit run app.py

# Lab 3: Pneumonia Classifier
cd Lab3_Model_Deployment
pip install -r requirements.txt
streamlit run app.py

# Lab 4: Flutter App
cd Lab4_Smart_App_Flutter/smart_app
flutter pub get
flutter run

Run Backend (Agentic AI)

# Start Ollama + Open WebUI
docker-compose up -d

# Start MCP Server
cd Control_Backend_AgenticAI
pip install -r requirements.txt
python main.py

Features

  • CNN Image Classification - Fashion item recognition with 95%+ accuracy
  • LSTM Time Series - Stock price prediction using RNN architecture
  • Model Deployment - Web interface for medical image analysis
  • Mobile App - Cross-platform Flutter app with Firebase authentication
  • Agentic AI - MCP-compliant backend with RAG and tool-calling

Tech Stack

Category Technologies
ML/DL TensorFlow, Keras, NumPy, Pandas, Scikit-learn
Web Streamlit, FastAPI
Mobile Flutter, Dart, Firebase
Backend Python, Ollama, MCP Protocol
DevOps Docker, Docker Compose

Contact

EMSI Engineering School - Computer Science
Website: emsi.ma


Built with Deep Learning and Mobile Development

About

A collection of labs in deep learning and Flutter: CNN-based pneumonia detection, LSTM stock forecasting, image classification, and a smart mobile app with AI features using TensorFlow, FastAPI, and local LLMs.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published