Shaofeng Zhang
Shaofeng Zhang
Yes, perhaps we will release the pretrained model weights in the next month.
See https://github.com/Kaiseem/QueryOTR/issues/9
Thanks for your reminder. This issue doesn't seem to matter because we use the fp16 computation provided by Accelerate.
See the eval_dir. The model will be automatically downloaded.
Running the evaluation command in the run.sh, and the model will be automatically downloaded.
Try **longer iterations** and larger batch sizes.
In the previous issue, we discovered that this checkpoint might not be successfully loaded during the evaluation process. This could be due to some structural errors that occurred when uploading...
You can follow the installation of U-ViT https://github.com/baofff/U-ViT.
Thank you for your attention to our work. However, as I cannot access the original server, I currently don't have sufficient computational resources to perform retraining. If you have adequate...
You have to train the model first, and load the trained weights.