zzzmm1

Will you plan to support group convolution?

1

Thanks for your good work！ Group conv is used in a lot of ConvNets. Will you plan to support group conv in the future version of dcn?

I am a little bit confused about the total sequence length of NOTU dataset. The four datasets of NFS100+OTB100+TC128+UAV123 have a total of 451 sequences. But the article mentions a...

I tried to replace the backbone or bottleneck with other convolutional layers with BN and ReLU. The training result is good but the validation and testing results are very low....

DualTFR paper

2

Dear authors: I wonder if your DualTFR paper is published? Could you please provide the address of the paper? I really want to read it. Thanks a lot!

When "pip install cuhnsw" in Ubuntu 16.04, meet a problem: ERROR: Could not build wheels for cuhnsw, which is required to install pyproject.toml-based projects Is there a built wheel file...

Save the search results like groundtruth_ivecs

11

I am not familiar with cuda programming. How can I minimally modify the code to save the search results like groundtruth_ivecs? Any help would be greatly appreciated! Thanks!

When the number of classes is very large, the loss function does not converge

1

In face recognition, when the number of classes exceeds 10 million and K=128, the CosFace/ArcFace loss function hardly converges, due to the fixed binary orthogonal codes. Any suggestions to help...

Datasets Preparation

3

Thanks for sharing your excellent work! I am a newcomer in the field of multi-label text classification. I don’t know where to download the train_data.json and test_data.json of Reuters-21578, also...

Could your provide raw results on all the mentioned datasets? Thanks!