🎆🎉 Object_detection-End-to-End-project 🎉🎆

Project Overview

• Objective: Develop an object detection system using the YOLOv5s pretrained model.

• Dataset:

  • Total Images: 2,086 images
  • Training Images: 1,461
  • Validation Images: 417
  • Testing Images: 208
  • Annotation Tool: Data was annotated using Roboflow for efficient labeling.

• Model Training:

  • Framework: Utilized the YOLOv5s model, known for its speed and accuracy.
  • Environment: Training conducted on Google Colab with GPU support.
  • Epochs: The model was trained for 250 epochs, resulting in significant improvements in detection accuracy.
  • Performance: The trained model demonstrated great accuracy on the validation and testing datasets.

• Additional Information: Further details, including setup instructions and usage, can be found in the project repository.

🎋 Project Tree Structure 🎋

Object_Detection-End-to-End-project/
  ├── .github/
  │   └── workflows/
  │       └── main.yaml
  ├── .jenkins/
  │   └── Jenkinsfile
  ├── artifact/
  │   └── 10_05_2024_03_23_14 (or time Stamp)/
  │       ├── data_ingestion/
  │       │   ├── feature_store/
  │       │   │   ├── test
  │       │   │   ├── train
  │       │   │   ├── valid
  │       │   │   └── data.yaml
  │       └── SignLangData.Zip
  │       ├── data_validation/
  │       │   └── status.txt           
  │       └── model_trainer/
  │           └── best.pt
  ├── data/
  │   ├── .gitkeep
  │   └── inputImage.jpg
  ├── flowchart/
  ├── logs/
  ├── notebook/
  │   ├── Object_Detection_with_Yolov5.ipynb
  │   └── test.ipynb
  ├── scripts/
  │   ├── ec2_setup.sh
  │   └── jenkins.sh
  ├── SignLanguage or src/
  │   ├── components/
  │   │   ├── __pycache__/
  │   │   ├── __init__.py
  │   │   ├── data_ingestion.py
  │   │   ├── data_validation.py
  │   │   ├── model_pusher.py
  │   │   └── model_trainer.py
  │   ├── configuration/
  │   │   ├── __pycache__/
  │   │   ├── logs/
  │   │   ├── __init__.py
  │   │   ├── gdown_connection.py
  │   │   └── s3_operations.py
  │   ├── constants/
  │   │   ├── __pycache__/
  │   │   ├── __init__.py
  │   │   └── training_pipeline/
  │   └── entity/
  │       ├── __pycache__/
  │       ├── __init__.py
  │       ├── artifact_entity.py
  │       └── config_entity.py
  ├── templates/
  │   └── index.html
  ├── yolov5/
  ├── app.py
  ├── data_collector.py
  ├── docker_compose.yml
  ├── Dockerfile
  ├── README.md
  ├── requirements.txt
  ├── setup.py
  └── template.py

Installation Instructions

1. Clone the repository:

   git clone https://github.com/shaheennabi/Object_detection-End-to-End-project.git
   cd Object_detection-End-to-End-project

• Create a new conda environment and activate it:

conda create -n signLanguage python=3.10 -y

conda activate signlanguage

• Install the required dependencies:

pip install -r requirements.txt

Export the environment variable(anaconda prompt)

set DATA_DOWNLOAD_URL=https://drive.google.com/uc?/export=download&id=  your file_id goes here"

Export the environment variable(git bash)

export DATA_DOWNLOAD_URL="https://drive.google.com/file/d/ your file_id goes here"

🎉 Workflow 🎉

After creating the project template:

• 🎇 Update constants 🎆
• 🎇 Update Entity modules 🎆
• 🎇 Update respective component 🎆
• 🎇 Update pipeline 🎆
• 🎉 Endpoint - (app.py) 🎉

🎆 Model Training 🎆

1. Ensure the dataset is properly loaded in the data/ folder.

2. Run the training script:

   python  cd yolov5/ && python train.py --img 416 --batch 16 --epochs 300 --data '../data.yaml' --cfg ./models/custom_yolov5s.yaml --weights 'yolov5s.pt' --name yolov5s_results  --cache

To use the trained model for inference on new images, run:

python cd yolov5/ && python detect.py --weights /content/runs/train/yolov5s_results/weights/best.pt --img 416 --conf 0.5 --source /content/test/images

Deployment

To deploy the application locally:

1. Install the dependencies in requirements.txt.
2. Run the Flask app:

   python app.py
   
   

Deployment (CI/CD)

Run these commands on EC2 Machine

• Update package lists

sudo apt update

• Update package lists (alternate command)

sudo apt-get update

• Upgrade the system

sudo apt upgrade -y

• Download Docker installation script

curl -fsSL https://get.docker.com -o get-docker.sh

• Run the Docker installation script

sudo sh get-docker.sh

• Add current user to the Docker group (run both lines)

sudo usermod -aG docker $USER
sudo usermod -aG docker $USER

• Install AWS CLI

sudo apt install awscli -y  (if not worked- install with curl -check doc)

• AWS Configuration

aws configure

Run these commands to install Jenkins on EC2

• Update package lists

sudo apt update

• Install OpenJDK 8

sudo apt install openjdk-8-jdk -y

# Jenkins repository (links for reference)
# https://pkg.jenkins.io/
# https://pkg.jenkins.io/debian-stable/

• Start Jenkins service

sudo systemctl start jenkins

• Enable Jenkins to start on boot

sudo systemctl enable jenkins

• Check the status of Jenkins service

sudo systemctl status jenkins

• Download Docker installation script

curl -fsSL https://get.docker.com -o get-docker.sh

• Run the Docker installation script

sudo sh get-docker.sh

• Add current user to Docker group

sudo usermod -aG docker $USER

• Add Jenkins user to Docker group

sudo usermod -aG docker jenkins

• Activate Docker group changes

newgrp docker

• Install AWS CLI

sudo apt install awscli -y

• Add Jenkins to Docker group again to ensure it has access

sudo usermod -a -G docker jenkins

• Download Docker installation script

curl -fsSL https://get.docker.com -o get-docker.sh

• Run the Docker installation script

sudo sh get-docker.sh

• Add current user to Docker group

sudo usermod -aG docker $USER

• Add Jenkins user to Docker group

sudo usermod -aG docker jenkins

• Activate Docker group changes

newgrp docker

• Install AWS CLI

sudo apt install awscli -y

• Add Jenkins to Docker group again to ensure it has access

sudo usermod -a -G docker jenkins

• Retrieve the initial Jenkins admin passwo

sudo cat /var/lib/jenkins/secrets/initialAdminPassword