search

LiJunnan1992 / DivideMix

Code for paper: DivideMix: Learning with Noisy Labels as Semi-supervised Learning
MIT License
529 stars 83 forks source link

Evaluation 50 classes WebVision #6

Closed DiegoOrtego closed 4 years ago

DiegoOrtego commented 4 years ago

Hi,

Many thanks for sharing the code of your cool work!

I am trying to evaluate your approach in the ImageNet val set when training with first 50 classes of WebVision. I wonder how do you get the right 50 classes from ImageNet to map the first 50 classes of WebVision.

LiJunnan1992 commented 4 years ago

Hi,

You can find the class sequence numbers (e.g. n01440764) in "synsets.txt" provided by webvision dataset, and use that to find corresponding imagenet classes.

DiegoOrtego commented 4 years ago

Cheers! All I needed! Btw, I am glad that our work (Unsupervised Label Noise Modeling and Loss Correction) inspired yours to drive our mixture modelling idea forward! Really cool a dynamic semi-supervised learning recomputing the clean subset based on label noise modelling with cross-network models ;)

LiJunnan1992 commented 4 years ago

Thanks! Your work was a big inspiration!