This repository was created as part of a project to explore the development of AI models for fashion style recommendations. It includes the implementation of machine learning and deep learning models from scratch, focusing on evaluating their performance and accuracy in generating cohesive fashion outfit recommendations.
The primary goal of the project was to compare the results of different models, including approaches inspired by GPT architecture, and assess their suitability for use cases in the fashion industry. The repository serves as a learning resource for understanding the application of AI in style pairing and recommendation systems.
Key features of this repository include:
- Implementation of multiple models for style recommendation.
- Benchmarking and comparison of model performance and accuracy.
- Exploration of AI-driven solutions for creating cohesive fashion styles.
This work combines foundational AI principles with real-world applications in the fashion domain. It is ideal for those interested in machine learning, deep learning, and AI applications in fashion technology.