search

roatienza / deep-text-recognition-benchmark

PyTorch code of my ICDAR 2021 paper Vision Transformer for Fast and Efficient Scene Text Recognition (ViTSTR)
Apache License 2.0
284 stars 57 forks source link

About the parameter `--valid_data` in the training command mentioned in README.md #16

Closed lexiaoyuan closed 2 years ago

lexiaoyuan commented 2 years ago

Hi, thanks for your work! When training, should the parameter --valid_data in the command be followed by data_lmdb_release/validation? But I found it written as data_lmdb_release/evaluation in README.md. Looking forward to your reply!

roatienza commented 2 years ago

In ViTSTR, the evaluation split was used to choose the best performing model. So, that was argument that follows --valid-data . The alternative is to use validation split but that will not necessarily find the best performing model on the target test dataset.