Application of Machine Learning and AI methods, such as Convolutional Neural Networks (CNN) and some preprocessing techniques for building a model capable of recoginizing manuscript digits precisely (PA 4 from Artificial Intelligence Course - DCOMP - UFSJ).
-
Python3 and pip package manager:
sudo apt install python3 python3-pip build-essential python3-dev -
virtualenv tool:
pip install virtualenv -
Libraries: Keras, TensorFlow, scikit-learn, Matplotlib, numpy, gdown and OpenCV;
-
Environments: Jupyter.
-
Clone the repository
git clone https://github.com/juliorodrigues07/manuscript_digit_recognition.git -
Enter the repository's directory
cd manuscript_digit_recognition -
Create a virtual environment
python3 -m venv .venv -
Activate the virtual environment
source .venv/bin/activate -
Install the dependencies
pip install -r requirements.txt
-
To visualize the notebook online and run it (Google Colaboratory), click here:
-
To run the notebook locally, run the following command in the notebooks directory:
jupyter notebook digit_recognition.ipynb
.
├── README.md # Project's documentation
├── requirements.txt # File containing all the required dependencies to run the project
├── plots # Directory containing all the graph plots generated
├── docs
| ├── Documentação.pdf <- Detailed documentation about the project
| └── LaTeX
├── notebooks # Directory containing project's main jupyter notebook
| └── digit_recognition.ipynb
├── datasets # Directory containing all used or generated datasets in the project
| └── digits # Directory containing own made digits for testing models
└── models # Directory containing all generated models in the project
-
To uninstall all dependencies, run the following command:
pip uninstall -r requirements.txt -y -
To deactivate the virtual environment, run the following command:
deactivate