hikopensource/DAVAR-Lab-OCR

(lgpma) inference with big input shape

Opened this issue · 1 comments

Congratulate your works,
I have a custom dataset with big shape image so I want to retrain and inference with bigger size. Could you show me how to config size when training and using bigger size in inference?
Thank you very much!

You may customize the input image scale by modifing the parameter in data pipeline. See 'demo/table_recognition/lgpma/configs/lgpma_base.py' (L164 controls the training scales, L182 controls the testing scales). The module implementation is in 'davarocr/davar_common/datasets/pipelines/transforms.py'