Nolan

Thanks for the wonderful job! I wonder how many epoches have you trained on the SCANNET dataset, 1024 epochs?

Setting of β

1

Hi. In the paper, the authors said "As for parameter β in eq. 2, it usually varies about 0.1, as we set it to 10^3 divided by number of elements...

"NoneType object is not callable" when running the train script

1

File "create_dataset_and_use_readerop.py", line 79, in reader_tensors = multi_vi_h5_data_reader(len(data_tensors_keys), json.dumps(reader_params)) TypeError: 'NoneType' object is not callable

Thanks for your sharing! I wonder hwo you determine the final results. Were you using the model trained in the last epoch to test or test in each epoch and...

Thnaks for sharing the code. Could the code reproduce the perfromance on Imagenet? Besides, how you train EfficientNetv2 to get the performance?

这个版本能否直接提交基金委

7

程老师及各位同行好！Latex编排的确实比word好看，只是不知道是否有人将这个正文部分pdf提交基金委系统，是否会出现格式不通过的问题？

Questions about FeedforwardController

1

Hello. This is a very good code. But I am very curious about why use FeedforwardController instead of the original BaseController. Is there small specific function?

Thanks for the code. However I am very confused with the parameter dataset_sampling with options: active_learning and naive, it is about the random sampling? Besides, the parameters are both in...

The results will decrease largely when I adopt 8 GPUs. How should I tune the learning rate.

Extend to batch SVD

1

It is very interesting. But I wonder whether it could be easily extended to handle batch SVD? Thanks!