Doodle Recognition System (DRS) AI Models

This repository contains machine learning notebook DRS_model.ipynb dedicated to the development and evaluation of AI models for the Doodle Recognition System (DRS).

About DRS_model.ipynb

The DRS_model.ipynb is a Jupyter notebook that includes the following components:

Data Loading and Preprocessing

• Data Loading: The notebook begins by loading a dataset comprising 345 classes of doodles.
• Preprocessing: It includes preprocessing steps that are crucial for preparing the doodle data for training the models.

Model Development

• CNN Model: Development of a Convolutional Neural Network (CNN) for recognizing and classifying doodle images.
• RNN Model: Development of a Recurrent Neural Network (RNN) that focuses on the sequential nature of drawing data.

Training

• Both models are trained on the dataset to learn the classification of doodles into one of the 345 categories.

Performance Evaluation

• Accuracy: Calculation of how often the models correctly predict the class of the doodles.
• F1 Score (Macro): A measure that balances the precision and recall of the models across all classes.
• AUC (Area Under the Curve): From the Receiver Operating Characteristic (ROC) curve, assessing the models' ability to discriminate between classes.