You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I would like to ask you a few questions, thank you.1.Can I use my own dataset for training? If it's possible, do I just need to prepare my own dataset and modify the config, or are there other modifications required?
2.For a custom dataset, do I only need images and mask images?
The text was updated successfully, but these errors were encountered:
@shiyuancao Thanks for your interest in our work. The answers to your questions are:
(1) You can train the model on your own dataset. You just need to prepare your data similar to the instructions in './readme/prepare_data.md'. Then you need to modify the 'dataloader' namespace in a config file.
(2) For a custom dataset, you only need images and mask images. In factor, the mask images are optional. The scripts can generate random masks with proper configurations.
I would like to ask you a few questions, thank you.1.Can I use my own dataset for training? If it's possible, do I just need to prepare my own dataset and modify the config, or are there other modifications required?
2.For a custom dataset, do I only need images and mask images?
The text was updated successfully, but these errors were encountered: