Chaoyang He

icon indicating a website link http://chaoyanghe.com icon indicating an email [email protected]

icon company FedML, USC icon location California icon location Co-founder at TensorOpera AI, FedML, and ChainOpera AI; Ph.D. in Computer Science, University of Southern California, Los Angeles

Results 162 comments of Chaoyang He

New paper

@Alan-Qin and @anqigao Please "Pull requests" your suggestion directly, thanks.

Issue when running centralized/bilstm_exps/main_text_classification.py

Hi! We at FedML have launched a new platform for FedNLP where there should be no such issue. Can you please check whether you face the same issue there? Here...

Issues on FedScale and Oort: (1) widely promotes its so-called advantages that are not based new version of FedML (10 months outdated until today) (2) The evaluation of FedML old version (Oct, 2021) is not based on facts and overlapping with a published paper; (3) unrealistic overlap between system efficiency and data distribution (4) issues on numerical optimization (5) dual submission?

@fanlai0990 Please do not close this issue before addressing the concern. I am waiting for your paper update to clarify our concern. FedML team, USC lab, and many collaborators take...

Issues on FedScale and Oort: (1) widely promotes its so-called advantages that are not based new version of FedML (10 months outdated until today) (2) The evaluation of FedML old version (Oct, 2021) is not based on facts and overlapping with a published paper; (3) unrealistic overlap between system efficiency and data distribution (4) issues on numerical optimization (5) dual submission?

This is my final reply. I am happy to see more researchers working on federated/distributed ML. Your contribution definitely pushes forwards the progress in this domain. You will earn more...

No such file or directory: 'cache_dir/distilbert_distilbert-base-uncased_cached_256_ClassificationModel_20news_niid_label_clients=100_alpha=5.0_-1'

Hi @DeviRule Please check this data loading issue.

Accuracy and loss didn't improve for FedAvg on 20news

@haofuml I see. I will check this issue this weekend.

Accuracy and loss didn't improve for FedAvg on 20news

@shubham-malaviya set the optimizer to "fedOPT", it should work

Cannot load "ciao" and "epinions" datasets

@emirceyani

The distributed experiment was stuck after creating model done

@csshali @csshali Thank you for the feedback. I've fixed this issue. Please try to update the latest source code.: https://github.com/FedML-AI/FedGraphNN/commit/5b3c766baa48efdde7aec658eaabb6d0eb19386f

Object Detection Task

@alpereninci @czkyx Yolov5 is always at yolov5 branch: https://github.com/FedML-AI/FedCV/tree/yolov5