README.md

Machine Learning Assisted Dual Harmonic Generation FROG for Enhanced Ultrafast Pulse Recovery

Wallace Jaffray¹, Ziheng Guo², Andrea di Falco², Marcello Ferrera^1*

¹ Heriot Watt University, Edinburgh, EH14 4AS, United Kingdom

² School of Physics and Astronomy, University of St. Andrews, Fife, KY16 9SS, United Kingdom

---

This is the collaboration project for ultrafast laser pulse response prediction between Heriot-Watt University and the University of St Andrews

This is the mock code for the paper. During to the shape of FROG traces are easy regonized, thus using the encoder-decoder model performs roughly the same as using Resnet9. The code is shared so people can do similar projects without trouble. if you like our project, please reference our paper.

How to use:

Dataholder has single input and dual input dataset.

Train Single only uses one inpout

Train Dual uses two input

Prediction Dual Dataset predicts pulse from dataset selection

Prediction Dual Specific predicts pulse from selected data.