Overview

From scratch implementation of CLiP

Technical Details

• Framework: PyTorch
• Dataset: Flickr
• Category: Computer Vision

Implementation Details

Learning Transferable Visual Models From Natural Language Supervision

Datasets

flickr 8000: Link

Frameworks:

Pytorch

Results (on T4 GPU Single)

Training epochs: 30

Train loss: 1.3 Val loss: 2.2

Source Code

📁 GitHub Repository: CLiP

View the complete implementation, training scripts, and documentation on GitHub.