Dataset for the Manuscript: Demonstration of optically-driven plasmonic nanomotors designed by deep learning networks

This repository contains the data corresponding to the manuscript "Demonstration of the optically-driven plasmonic nanomotor designed by deep learning networks." It consists of 5 parts: Machine learning, Numerical analysis, Rotation measurement, Scattering measurement, and Supplementary information. The code for the machine learning algorithm is available at "https://github.com/mintaechung/Nanomotor_Predictor_Generator."

'Machine_learning.zip': Correlation between optical torques calculated by SIE and predicted by trained CNN, Objective loss functions at the 1st iteration, and the torque distribution of the initial randomset and the output of the nanorotor generator after the 3rd iteration.

'Numerical_analysis.zip': MATLAB codes to retrieve 'Moments', 'Field intensity distribution', 'Poynting vectors', and 'Torques'. 

'Rotation_measurement.zip': Raw videos, Intensity profiles of ROI, Rotation measurement results.

'Scattering_measurement.zip': Scattering intensity measurement with reference light.

'Supplementary_Info.zip': Random geometry generation, Optical torques of 6 blades, Expanded structure, Shrinkage, Polarization independence, Angular momentum, and Machine learning progress.