search

shipra25jain / ESSNet

Embedding-based Scalable Segmentation Network
MIT License
28 stars 6 forks source link

About the training settings on COCO+LVIS #6

Closed hsfzxjy closed 3 years ago

hsfzxjy commented 3 years ago

Hi,

I am trying to train ESSNet on COCO+LVIS, but seemingly could not find the settings of some parameters (B, k, d) in your paper. Could you give some advice?

Thank you.

shipra25jain commented 3 years ago

@hsfzxjy , we will update the arxiv file this week, with additional information for COCO+LVIS experiments.

Thanks

shipra25jain commented 3 years ago

@hsfzxjy , we have now updated our arxiv paper at https://arxiv.org/pdf/2012.07489.pdf You will find more details about COCO+LVIS dataset. Also, let me know if you have any doubt regarding the building of COCO+LVIS dataset.

hsfzxjy commented 3 years ago

@shipra25jain Thank you for your clarification! The settings in paper is clear enough and I will try them in the following days.