README.md

# My CLIP Video-Text Model

This model was trained on the MSR-VTT dataset using a custom CLIP-based architecture.
Now using an N-pairs margin loss for training.