Hsin-Ping Chou comments

Results 5 comments of


                                            Hsin-Ping Chou

I have read the enas paper，which says that weight sharing is used for efficiency. Any one can tell me which part of the code realize the weight sharing?Thanks a lot!

Hi, ENAS builds a DAG to achieve weight sharing. The DAG is created in this file. [https://github.com/melodyguan/enas/blob/master/src/cifar10/general_child.py](url)

Points for sampler

I think you're right. I also found this issue when running the code. I think it only effects "Resample" and can be fixed by adding a line after the "ImbalancedDatasetSampler"...

Points for sampler

You should probably wait for the author's response before closing the issue though I think you're right about the question. And it's also better to keep the issue opened before...

ERM Baseline

Hi, I think there's nothing wrong with the ERM baseline. You're getting this "one" better results due to the seed choice for numpy which efftects how the imbalanced data is...

Graphs of accuracy and loss during final training process

Hi, I suppose you're showing the results of running micro_final.sh or macro_final.sh. This phenomenon is due to the cosine learning rate scheduling with restart.