search

GeorgeCazenavette / mtt-distillation

Official code for our CVPR '22 paper "Dataset Distillation by Matching Training Trajectories"
https://georgecazenavette.github.io/mtt-distillation/
Other
389 stars 54 forks source link

Checkpoints of models #38

Open DeepOceanDeep opened 1 year ago

DeepOceanDeep commented 1 year ago

Good day @GeorgeCazenavette I was reading this "Distillation by Matching Training Trajectories" paper and a more recent paper of yours, and looked into the both repos. I was wondering if by any chance you have stored the checkpoint of models that are Trained from scratch on the whole dataset? Like the Table 8 model checkpoints form the GLaD paper?
Thanks in advance. Flora

Tankiit commented 6 months ago

Hi I too am wondering if you managed to store checkpoints which could be really helpful to kind of see if we could utilise for other downstream tasks ? Regards Tanmoy